Standing on Principles Collected Essays

S T AND ING ON PR INC I P L E S

This page intentionally left blank

STANDING ONPRINCIPLES

C O L L E C T E D E S S A Y S

TOM L . B EAUCHAMP

12010

1Oxford University Press, Inc., publishes works that further

Oxford University’s objective of excellence

in research, scholarship, and education.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi

Kuala Lumpur Madrid Melbourne Mexico City Nairobi

New Delhi Shanghai Taipei Toronto

With offices in

Argentina Austria Brazil Chile Czech Republic France Greece

Guatemala Hungary Italy Japan Poland Portugal Singapore

South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright � 2010 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016

www.oup.com

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,

electronic, mechanical, photocopying, recording, or otherwise,

without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

Beauchamp, Tom L.

Standing on principles : collected essays / Tom L. Beauchamp.

p. ; cm.

Contains articles published previously in various sources.

Includes bibliographical references and index.

ISBN 978-0-19-973718-5

1. Bioethics. 2. Human experimentation in medicine—Moral and ethical aspects. I. Title.

[DNLM: 1. Bioethics—Collected Works.

2. Human Experimentation—Collected Works. WB 60 B371s 2010]

QH332.B425 2010

1740.957—dc22

2009033282

1 3 5 7 9 8 6 4 2

Printed in the United States of America

on acid-free paper

www.oup.com

To

Ruth R. Faden

Partner for life and a magnificent human being in every way of being human.

CONTENTS

Acknowledgments ix

Introduction xiii

PART I The Belmont Report and the Rise of Principles

ONE The Origins and Evolution of the Belmont Report 3

TWO Codes, Declarations, and Other Ethical Guidance for Human

Subjects Research: The Belmont Report 18

PART II Principlism and Practice

THREE The Four Principles Approach to Health Care Ethics 35

FOUR Informed Consent: Its History and Meaning 50

FIVE Who Deserves Autonomy and Whose Autonomy Deserves

Respect? 79

SIX The Concept of Paternalism in Biomedical Ethics 101

SEVEN When Hastened Death is Neither Killing nor

Letting Die 120

EIGHT The Exploitation of the Economically Disadvantaged in

Pharmaceutical Research 132

PART III Theory and Method

NINE Principles and Other Emerging Paradigms

in Bioethics 153

TEN A Defense of the Common Morality 175

ELEVEN From Morality to Common Morality 189

TWELVE On Eliminating the Distinction Between Applied Ethics and

Ethical Theory 211

THIRTEEN Does Ethical Theory Have a Future in Bioethics? 229

FOURTEEN The Failure of Theories of Personhood 247

FIFTEEN Looking Back and Judging Our Predecessors 261

Index 281

viii C O N T E N T S

ACKNOWLEDGMENTS

Many parts of this book are connected to my work with James Childress. Our

collaboration has been a two-way conversation, and he has been an abiding help

in stimulating my thought on several of the essays in this volume. In 1978, Jim and

I decided on the persons to whom we would dedicate the volume. On my side, I

dedicated the volume to Ruth Faden and Donald Seldin, with whom I was in

frequent conversation at the time about the embryonic but developing field of

bioethics. I chose well. Now, more than 30 years later, they remain the deepest

influences on my work on the subjects in this volume. Their personal dedication

to me and their professional facilitation of my work have been remarkable.Wayne

Davis, my department chair, deserves a similar acknowledgment for his support

over the last 20 years. Perhaps no one is a perfect chair, but Wayne seems so, and

he is a model of the honest philosopher.

Philosophers should learn from their critics, and I have had some dedicated

critics over the years on several of the subjects addressed in this volume. I

acknowledge in particular the many contributions made by John Arras, Bernard

Gert, Danner Clouser, Albert Jonsen, Ronald Lindsay, Edmund Pellegrino, and

David DeGrazia. All have offered trenchant criticisms. Other valued criticism has

come from Oliver Rauprich, Rebecca Kukla, Carson Strong, Peter Herissone-

Kelly, Avi Craimer, Frank Miller, and Robert Veatch.

Essay 1, “The Origins and Evolution of the Belmont Report,” and Essay 2,

“Codes, Declarations, and Other Ethical Guidance for Human Subjects

Research: The Belmont Report,” are centered on the years I worked for the

National Commission for the Protection of Human Subjects (1975–1978), some

of the most rewarding years of my life. I owe thanks to the many people who

supported me in writing for the Commission, especially the writing of Belmont.

I owe special acknowledgments to Patricia King, Michael Yesley, Donald

Seldin, Albert Jonsen, Stephen Toulmin, Robert Levine, Barbara Mishkin, and

Kenneth Ryan—and, for later criticisms, Ernest Marshall. They all caused me

to rethink my ideas many times over. I am also grateful to Jay Katz, Samuel

Gorovitz, LeRoy Walters, Tris Engelhardt, Charles McCarthy, and John

Robertson for conversations and help in research during these years. For

excellent editorial advice while writing Essay 2, I recognize the contribution

made by Zeke Emanuel, Christine Grady, Robert A. Crouch, Reidar Lie, Frank

Miller, and David Wendler, as editors of the book in which the piece appeared.

Frank is owed special thanks for his efficient pursuit of my piece.

Essay 3, “The Four-Principles Approach to Health Care Ethics,” and Essay 9,

“Principles and Other Emerging Paradigms in Bioethics,” are centered on prin-

ciplism. The first of the two essays was written because Raanan Gillon persuaded

me to compose it and then helped me make it as good as it could be. Raanan has

been supportive throughout my career, and I thank him for contributions that

reach far beyond this essay. I also owe Ed Pellegrino an acknowledgment for his

many critiques of the four-principles approach, including a critique when I wrote

this essay. My thinking on these issues has been pushed forward by some stimu-

lating criticisms, published and unpublished, by Bernard Gert, Danner Clouser,

Charles Culver, John Arras, David Smith, Alasdair Campbell, John Harris, and

Dan Callahan. On the constructive side, I have enormously benefited from the

work of my colleague Henry Richardson on the subject of specification. Henry has

also been a supportive critic.

Essay 4, on “Informed Consent: Its History and Meaning,” appeared because

Bob Veatch persuaded me to write it and offered sound editorial advice during its

drafting. Ruth Faden stimulated me to my best ideas on informed consent, and

much of what I say here derives from our 1986 book, A History and Theory of

Informed Consent. When writing that book we received wonderful support and

criticism from Bettina Schone-Seifert and Nancy King; and we were inspired at

the time by the published work of, and personal consultation with, Jay Katz, Allen

Buchanan, Robert Levine, Alan Meisel, and Sissela Bok.

Essay 5, “Who Deserves Autonomy and Whose Autonomy Deserves Respect?,”

came into being because I was persuaded by James Stacey Taylor to rethink and

deepen my previous publications on autonomy. On the matters of autonomy

treated in this essay, many people have contributed to my thinking. Almost

certainly, the deepest influences have come from work with Ruth Faden and

through discussions with Joel Feinberg, who was working on his theory of

autonomy as I was first developing mine. I still today regard Feinberg as the

soundest writer on many of the subjects I address. In what seems a lifetime ago, I

had several discussions with Gerald Dworkin that helped me get my thoughts

straight on various issues about autonomy.

Essay 6, “The Concept of Paternalism in Biomedical Ethics,” is the final phase, I

hope, in my long history of probing the subject of paternalism. Jim Childress and I

have struggled together through the muddy conceptual issues in this area.

Although we have never come to complete agreement, Jim has convinced me to

moderate my skeptical views about soft paternalism. Jim and Ruth Faden have

been instrumental in helping me get my thoughts straight. In the early years of my

x A C K N OW L E D G M E N T S

thinking about the key problems (dating to 1975), I was influenced by Joel

Feinberg, with whom I had stimulating conversations on the subject. My work

stands in the shadow of Feinberg’s seminal writing on the subject.

Essay 7, “When Hastened Death Is neither Killing nor Letting Die, brings together

much ofmy thinking about physician-hastened dying. TimQuill and Peg Battin were

my editors and advisors when I wrote this essay. The compactness and directness of

the argument in this essay owe much to them. I’ve learned a great deal about what to

say and not say about these issues from conversations with, and the publications of,

Dan Brock and James Rachels. I have had several public debates with Ed Pellegrino on

the topic. Ed has many times helped me understand views that I don’t hold; he has a

wonderful ability to make them clear and plausible. In the early years in which my

views on these subjects were developing, I acknowledge how much I enjoyed

examining ideas through conversations with Arnold Davidson. Every philosopher

should have as much fun as Arnold and I had in those years.

Essay 8, “The Exploitation of the Economically Disadvantaged in

Pharmaceutical Research,” was originally written for a conference. I thank the

organizer of the conference, Denis Arnold. At the conference, I received useful

criticism from Dan Callahan, Norman Daniels, and Dan Wikler. My work on

pharmaceutical research in all its dimensions has been greatly facilitated by many

years of discussing these issues with Robert Levine. In working on the subject of

global justice and the economically disadvantaged, I have had stimulating inter-

actions with Ruth Faden, Madison Powers, and Thomas Pogge.

Essay 10, “A Defense of the Common Morality,” and Essay 11, “From Morality

to Common Morality,” are devoted to the subject of common morality. As

anyone can see by reading the second of the two essays, I owe an immense

amount to Bernard Gert, not only for his criticisms of me, but for the constructive

side of his moral theory. It is a pleasure to be able to acknowledge the influence of

his bold criticisms and his constructive theories. My essay was written for and

delivered at Dartmouth College on the occasion of Gert’s retirement and 50th year

of teaching at Dartmouth. I was helped in formulating my topic by Jim Moor,

Bernie’s colleague at Dartmouth. Dan Brock, Ron Green, and Don Marquis gave

me stimulating criticisms during the visit to Dartmouth that helped me redraft.

More generally on the subjects of commonmorality and principlism, much in my

work has been either motivated by or improved by the work of other critics,

especially David DeGrazia, Carson Strong, and Ronald Lindsay.

Essay 12, “On Eliminating the Distinction Between Applied Ethics and Ethical

Theory,” and Essay 13, “Does Ethical Theory Have a Future in Bioethics?,” are two

of my forays into the waters of ethical theory and its limitations. The first, Essay 12,

was written long ago; it was invited by the two editors of theMonist in those years,

Eugene Freeman and John Hospers. Not much had been published on this subject

at the time, but I did greatly benefit by Gert’s early work on the subject. I owe the

A C K N OW L E D G M E N T S xi

idea behind Essay 13, and criticism of it, to Jeff Kahn and Anna Mastroianni, the

two editors of the journal issue in which this essay appeared.

Essay 14, “The Failure of Theories of Personhood,” was first delivered as an

address in Hong Kong, where I had the good fortune to receive excellent criticism

from Gerhold Becker, John Harris, and Michael Quante. Later I received equally

helpful criticisms fromRobert Veatch andHenry Richardson when presenting the

paper at Georgetown. Throughout the period of my writing this essay, I had many

discussions about the issues with Raymond Frey.

Essay 15, “Looking Back and Judging Our Predecessors,” was published in the

form of an exchange between Allen Buchanan and me. Allen and I largely agreed

on every major issue, but I was given a distinct advantage when writing my essay:

Allen sent his draft first, and I was able to learn a good bit from him. I owe much

in my presentation to the structure he laid out in his essay, and both of us owe

much to the work of the Advisory Committee on Human Radiation Experiments

(ACHRE), Final Report of the Advisory Committee on Human Radiation

Experiments (New York: Oxford University Press, 1996), on whose conclusions

Allen and I were commenting. In getting straight on the history surrounding the

radiation experiments, I learned a great deal from Ruth Faden, Jonathan Moreno,

and Susan Lederer.

In the Introduction to this volume, I have supplied the names of publishers and

full bibliographical and copyright information pertaining to all original sources of

these Collected Essays.

xii A C K N OW L E D GM E N T S

I NTRODUCT ION

This volume collects essays and treatises on several subjects that I have written

over the course of 25 years. All are on philosophical and moral issues in the field of

biomedical ethics. The topics range from the historical origins of modern research

ethics to substantive issues in bioethics about moral principles and methodology.

Despite the diversity of topics, a specific unity holds the three parts of this collection

together. The unifying theme is the transparent connection of these essays to many

of the topics and chapters in Principles of Biomedical Ethics (hereafter Principles),

which I coauthored with James Childress. All readers familiar with the basic

structure of that book will see that these Collected Essays augment, develop, and

defend some of its central positions and arguments. A few essays take off in new

directions, but all have a connection to themes in Principles. I have tried to select only

essays that expand and deepen, while not duplicating, material in Principles.

This introduction explains the publishing history and content of these essays, as

well as ways in which they augment and develop the so-called “principlist” theory

that Childress and I developed together. The year, source, and place of publication

and an abstract of each essay are supplied.

The book is divided into three parts. The first part is entitled “The Belmont

Report and the Rise of Principles.” The two essays in this part explain the emerging

importance of frameworks of principles in bioethics in the mid-1970s. The

primary sources at the time of the rise of these frameworks were Principles and

the Belmont Report. The latter is a government report that I drafted for the

National Commission for the Protection of Human Subjects of Biomedical and

Behavioral Research. The history of the writing of both of these works and their

crossover influence are explained and analyzed in Part I.

The second part is entitled “Principlism and Practice.” This section treats

several issues about moral principles and their practical use in bioethics as well

as the nature of “principlism.” This word was coined by Danner Clouser and

Bernard Gert principally to refer to the theory Childress and I developed. The first

essay in this second section adopts the language of “The Four-Principles

Approach,” an expression originally coined by Raanan Gillon to refer to the

theory in Principles. The material in Part II develops some themes Childress and

I address in Chapters 4 through 7 of Principles (using 6th edition chapter

numbers).

The third part concentrates on questions of theory and method in ethics—long

mymajor teaching and publishing interest in bioethics. This part treats the idea that

the four-principles approach provides a theoretical framework or paradigm for

bioethics togetherwith some ideas about theory andmethod that include discussion

of the nature and role of the commonmorality. This part also probeswhether ethical

theory has a strong role to play in the future of bioethics and whether so-called

“applied ethics” should be sharply distinguished from general ethical theory. Two

other essays investigatewhether the concept of person has a significant role to play in

bioethics and how judgments can be made of the actions and character of persons

who lived in past decades and centuries when they embraced significantly different

moral standards. The material in the third part develops answers to various

problems that Childress and I address in Chapters 1, 2, 3, 9, and 10 of Principles.

During the preparation of this volume I have lightly edited almost all of these

essays for purposes of clarity and style. I have added content to three of the essays.

This supplementary content has in every case been drawn from other parts of my

published work. In no case have I altered the basic structure or argument found in

the original publication. I have altered a title only in the case of the fourth essay:

“Informed Consent: Its History and Meaning”; it was originally published under

the less specific title “Informed Consent.”

I also explain below precisely what I have done to alter the few essays in which

material has been added in these Collected Essays.

PART I. THE BELMONT REPORT AND THE RISE OF PRINCIPLES

Essay 1. “The Origins and Evolution of the Belmont Report”

Publication Data. This essay was published in Belmont Revisited: Ethical Principles

for Research with Human Subjects. Copyright � 2005 by Georgetown University

Press. From Belmont Revisited: Ethical Principles for Research with Human Subjects,

James Childress, Eric Meslin, and Harold Shapiro, Editors, pp. 12–25. Reprinted

with permission. www.press.georgetown.edu. This essay, as here slightly revised,

also draws on a small body of material in an essay I wrote in The Story of Bioethics,

also published by Georgetown University Press. Reprinted with permission.

Abstract. This 2005 article recounts my work in delineating a framework of

principles of research ethics when I was in the position of consultant philosopher

at the National Commission for the Protection of Human Subjects of Biomedical

and Behavioral Research (U.S. Congress and NIH, 1974–78). The article starts with

the historical context in which principles rose to prominence in biomedical ethics.

xiv I N T R O D U C T I O N

www.press.georgetown.edu

I concentrate here on research ethics. The project of creating a framework of basic

principles for all federally funded research had been mandated by a public law

enacted by the U.S. Congress. I was at work on the volume with Childress when I

accepted an assignment with the National Commission to write its Belmont

Report. I explain both the history of my drafting of this Report, how it was revised

in the course of deliberations at the National Commission, and the connection of

this drafting to my concurrent work with Childress. I discuss how the National

Commission viewed these principles as embedded in preexisting public morality

and how it regarded its set of principles as a universally valid resource for the

formulation of public and institutional policies of research ethics. I outline how

the views in Belmont express the basic structure for research ethics as it developed

in the last quarter of the twentieth century and remains today.

Essay 2. “Codes, Declarations, and Other Ethical Guidance for Human Subjects

Research: The Belmont Report”

Publication Data. This essay was published in The Oxford Textbook of Clinical

Research Ethics, ed. Ezekiel Emanuel, Christine Grady, Robert Crouch, Reidar Lie,

Franklin Miller, and David Wendler (New York: Oxford University Press, 2008),

149–155. By permission of Oxford University Press (New York).

Abstract. This essay was commissioned by a group of scholars in bioethics in the

Department of Bioethics at the National Institutes of Health. It picks up the discus-

sionof theBelmontReport roughlywhere I ended the discussion in the first essay. This

second essay further examines the history of the Report together with an explanation

of the National Commission’s larger body of publications, which comprised

17 volumes. It discusses the moral content of the principles that were adopted and

the idea that the principles form a basic moral framework for research ethics.

The essay contains a critical evaluation of the philosophical roots of

the principles and includes a section that examines philosophical questions

about and weaknesses in the principles (as they are expressed in the Belmont

Report), including some weaknesses that persist still today in research ethics. The

essay concludes with a section on the influence and ongoing significance of the

Belmont Report, which still stands as an internationally influential government-

commission statement of moral requirements in research ethics. It is possibly

more widely known than any document in research ethics other than the

Declaration of Helsinki, which at present seems to be in a stage of declining

influence. Belmont was eventually adopted by all relevant agencies of the

U.S. government as a statement of the obligations scientific investigators must

discharge in conducting human research. I explain how Belmont has been one of

the few documents to have influenced almost every sphere of activity in

I N T R O D U C T I O N xv

bioethics: moral theory and general standards of professional ethics, government

regulatory activity, bioethics consultation, and even medical practice.

PART II. PRINCIPLISM AND PRACTICE

Essay 3. “The Four Principles Approach to Health Care Ethics”

Publication Data. This essay was published in Principles of Health Care Ethics, 2nd

ed., Richard Ashcroft, Angus Dawson, Heather Draper, and John Macmillan, eds.

(London: John Wylie, 2007), 3–10. By permission of Wiley-Blackwell, 9600

Garsington Road, Oxford OX4 2DQ.

Abstract. This essay was first published in the early 1990s with the goal of

explaining and critically examining the four principles or “principlist” account of

biomedical ethics to an international audience. The first edition of the Principles of

Health Care Ethics was the creation of British physician and medical ethics scholar

Raanan Gillon. This anthology was devoted primarily to critical appraisal of the

Principles book that Childress and I published. My essay on “The Four-Principles

Approach” was the opening essay in both the first and the second editions. (The

essay is published here as it was revised and updated for the second edition.) The

essay is a basic and reasonably comprehensive explanation of the four-principles

approach, clarifying various of its claims and attempting to straighten out assorted

misunderstandings of the Principles book. The following topics are discussed: the

origins of principled frameworks in bioethics, the nature of the framework that

Childress and I use, the centrality of the common morality in our work, the prima

facie character of principles and rules, the specification of principles and rules, and

the role of the method of coherence in moral justification.

Essay 4. “Informed Consent: Its History and Meaning”

Publication Data.This essay was published inMedical Ethics, ed. RobertM. Veatch

(Boston and London: Jones and Bartlett Publishers, Inc., 1st edition 1989; 2nd

edition, 1997), 186–205. Copyright� 1997 Jones and Bartlett Publishers, Sudbury,

MA. www.jbpub.com. Reprinted with permission. Some supplementary historical

material has been added that derives from the book I coauthored with Ruth Faden

entitled A History and Theory of Informed Consent (New York: Oxford University

Press, 1986). By permission of Oxford University Press.

Abstract. This essay is a comprehensive treatment of the history, nature, and

moral importance of informed consent. It begins by treating the near absence of

requirements of informed consent in the history of medicine. I trace how

xvi I N T R O D U C T I O N

www.jbpub.com

informed consent obligations and requirements gradually emerged from legal

cases, regulatory interventions, government-appointed commissions, and intra-

professional events in the last half of the twentieth century. Following this history,

I present an analysis of the concept of informed consent, including an analytical

treatment of its basic elements and conceptual conditions. This analysis might

seem to provide a solid foundation for a definition of “informed consent,” but I

argue that this term needs a deeper analysis. It must be understood in terms of two

common, entrenched, and irreducibly different meanings of “informed consent.”

I establish the two senses as (1) autonomous choice and (2) institutional consent. I

note that assessment of the quality of the consent is important for understanding

both senses of “informed consent” and for understanding requirements of

obtaining consent. I argue that the quality of consent should be judged by several

different considerations, including the level of understanding of disclosed infor-

mation and whether undue influence is present in making a request for consent.

Finally, I examine various justifications for waiving or at least not requiring

informed consent. Some are found to be valid reasons for not obtaining consent,

and others are found to be unjustified.

Essay 5. “Who Deserves Autonomy and Whose Autonomy Deserves Respect?”

Publication Data. This essay was published in Personal Autonomy: New Essays in

Personal Autonomy and Its Role in Contemporary Moral Philosophy, ed. James

Taylor (Cambridge: Cambridge University Press, 2005), 310–329. Copyright �2005 Cambridge University Press. Reprinted with the permission of Cambridge

University Press. A few paragraphs of supplementary material have been added

from my “Consent and Autonomy,” in Frank Miller and Roger Wertheimer, eds.,

The Ethics of Consent (New York: Oxford University Press, 2010). By permission of

Oxford University Press (New York).

Abstract. I distinguish between “autonomy,” “respect for autonomy,” and

“rights of autonomy.”Whereas “respect for autonomy” and “rights of autonomy”

are moral notions, “autonomy” and “autonomous person” are not obviously

moral notions. To some philosophers they seem metaphysical rather than

moral. However, this distinction between the metaphysical and the moral has

fostered precarious claims in moral philosophy such as these: (1) analysis of

autonomy is a conceptual and metaphysical project, not a moral one; (2) a

theory of autonomy should not be built on moral notions, but rather on a

theory of mind, self, or person; (3) the concept of autonomy is intimately

connected to the concept of person, which alone anchors the concept of moral

status. I assess each of these claims with the objective of determining who qualifies

as autonomous and which level (or degree) of autonomy deserves respect.

I N T R O D U C T I O N xvii

I argue that moral notions—in particular, respect for autonomy—should affect

how we construct theories of autonomous action and the autonomous person on

grounds that a theory of autonomy should be kept consistent with the substantive

assumptions about autonomy implicit in the principle of respect for autonomy.

However, theories of autonomy should only be constrained by the principle of

respect for autonomy, not wholly determined by it. I offer an abbreviated theory of

conceptually necessary conditions of autonomy. My conditions differ substan-

tially from prominent accounts in the literature, such as Harry Frankfurt’s theory,

which I criticize in section 4 of this paper. My discussion of the claim that the

concept of autonomy is intimately connected to the concept of person leads

directly to the content of essay 14 in these Collected Essays.

Essay 6. “The Concept of Paternalism in Biomedical Ethics”

Publication Data. This essay was published in the Jahrbuch fur Wissenschaft

und Ethik (2009). Berlin: Walter de Gruyter GmbH & Co. KG; Mies-van-

der-Rohe-Str. 1; 80807 Munchen, Germany. Permission by Rights & Licenses

Department.

Abstract. This essay pulls together in one place the several strands of my thought

over 35 years on the problem of paternalism (principally medical paternalism). In

the literature of biomedical ethics, paternalism has been both defended and

attacked in several areas of clinical medicine, public health, health policy, and

government agency policies (e.g., the policies of the Food and Drug

Administration). I argue that it is unclear in much of this literature what patern-

alism is and which types of paternalism, if any, are justified. The position closest to a

consensus position in the literature is that so-called strong (or hard) paternalism is

not justified, but I argue that strong paternalism can be justified and that it is the

only interesting and controversial form of paternalism. I try to show that questions

of the justification and the definition of paternalism are closely connected, and that

there is a definition that is grounded in fidelity to the principle of respect for

autonomy. I then discuss several of the most prominent practical problems about

paternalism in biomedical ethics. In each case I focus on how obligations of respect

for autonomy and beneficence need to be balanced whenmaking a judgment about

whether a paternalistic intervention is justified.

Essay 7. “When Hastened Death Is Neither Killing nor Letting Die”

Publication Data. This essay was published in Timothy E. Quill and Margaret

P. Battin, eds., Physician-Assisted Dying: The Case for Palliative Care and Patient

xviii I N T R O D U C T I O N

Choice, pp. 118–129. Copyright � 2004. The Johns Hopkins University Press.

Reprinted with the permission of the Johns Hopkins University Press.

Abstract. This essay covers a broad range of issues in what is today commonly

categorized as “physician-assisted suicide.” I start with a brief discussion of the recent

history of the issues, from the Quinlan case to the present. I then discuss issues about

rights of autonomous choice, includingwhether the capacity for autonomy is likely to

be reduced in circumstances of making a choice to hasten death and whether

so-called “coercive situations” sometimes deprive a person of autonomous choice.

I consider the justification of physician involvement in hastening death and whether

the physician’s intention makes a morally relevant difference. I also analyze concep-

tual features of the language generally used to discuss these subjects, including

“suicide,” “hastened death,” “killing,” and “letting die.” These concepts can make a

critical moral difference to how we think about whether an intentionally hastened

death constitutes either a suicide, a killing, or a letting die.

The meaning of these terms also can determine whether “suicide,” as used both

in medicine and beyond, entails disapproval and whether it is proper to use the

language of “causing death,” which suggests liability, when characterizing the

physician’s action of hastening death. Also assessed are whether “the right to die”

is a meaningful notion and, if so, how it differs from the right to refuse treatment,

the right to avoid suffering, and the right to death with dignity.

Essay 8. “The Exploitation of the Economically Disadvantaged in

Pharmaceutical Research”

Publication Data. This essay was published in Denis Arnold, ed., Ethics and the

Business of Biomedicine (Cambridge: Cambridge University Press, 2009).

Copyright � 2009 Cambridge University Press. Reprinted with the permission

of Cambridge University Press.

Abstract. I evaluate some searing criticisms of the power and influence of the

pharmaceutical industry that have appeared in various published sources. The

industry stands accused of a sea of injustices and corruptions, including aggressive

and deceptive marketing schemes, exploitative uses of research subjects, a cor-

rupting influence on universities, suppression of and amateurism in handling

scientific data, and conflicts of interest that bias research investigators. Each

charge of injustice derives from concern about some form of unfair influence

exerted by pharmaceutical companies. The array of alleged forms of influence is

vast, and I here telescope to one area: the recruitment and enrollment in clinical

research of vulnerable human subjects, in particular the economically disadvan-

taged. I focus on the charge that subjects in clinical trials are unjustly exploited by

manipulative and unfair payment schemes.

I N T R O D U C T I O N xix

I treat three problems. The first is whether the economically disadvantaged

constitute a vulnerable group. I argue that classification as a “group” is a misleading

characterization thatmay cause paternalistic overprotection. The second problem is

whether the vulnerable poor are exploited by payments that constitute either an

undue influence or an undue industry profit. I argue that such assessments should

be made situationally, not categorically. The third problem is whether the poor are

likely to give compromised or nonvoluntary consents. I argue that this third

problem, like the second, is subtle and complicated, but practically manageable,

and I add that pharmaceutical research involving the poor and vulnerable can, with

proper precaution, be carried out in an ethically responsible manner. Whether the

research is so conducted is another matter, an empirical problem beyond the scope

of my argument.

PART III. THEORY AND METHOD

Essay 9. “Principles and Other Emerging Paradigms in Bioethics”

Publication Data. Tom L. Beauchamp, as published in 69 Indiana Law Journal:

1–17 (1994). Copyright � 1994 by the Trustees of Indiana University. Reprinted

with permission. For the present volume I have added a few paragraphs to this

essay.

Abstract. This article is centered on several accounts of biomedical ethics

that allegedly challenge the principles account that Childress and I defend.

Leading critics of our principles (Bernard Gert, Danner Clouser, Albert

Jonsen, Stephen Toulmin, Carson Strong, John Arras, Edmund Pellegrino,

and others) have, since the late 1980s, defended some types of theory or

method proposed as alternatives to or substitutes for principlism. These types

include casuistry, virtue theory, and impartial rule theory. These accounts

were first emerging to prominence in bioethics at the time this article was

written. I welcome these developments in this essay, because they improve

the range, precision, and quality of thought in the field. I also acknowledge

the contribution those who embrace these paradigms have made to the

improvement of my own thought. However, I argue that impartial rule

theory, casuistry, and virtue ethics should not be presented as rivals to

principlism, because they neither replace the principles in principlism nor

are inconsistent with those principles. I argue that all leading “alternatives”

are compatible with, and not alternatives to, an approach based on princi-

ples. Finally, I point to some limitations of the principle-based approach, in

light of these paradigms, and reflect on how those limitations should be

handled.

xx I N T R O D U C T I O N

Essay 10. “A Defense of the Common Morality”

Publication Data. This essay was published in Kennedy Institute of Ethics Journal

13:3 (2003): 259–274. Copyright � 2003. The Johns Hopkins University Press.


Abstract. Phenomena of moral conflict and disagreement have led writers in

ethics to two antithetical conclusions: Either moral distinctions hold universally

or they hold only relative to a particular and contingent moral framework. If the

latter, they cannot claim universal validity. In this essay I defend a universalistic

account of the most general norms of morality in the course of responding to

some critics of the common morality theory that Childress and I defend in

Principles. In particular, I respond to criticisms by David DeGrazia and Leigh

Turner, both of whom take “common morality” to refer to a broader and quite

different body of norms than I do.

I maintain that one can consistently deny universality to some justified moral

norms while claiming universality for others. I argue that universality is located in

the common morality and that nonuniversality is to be expected in other parts of

the moral life, which I call “particular moralities.” The existence of universal

moral standards is defended in terms of (1) a theory of the objectives of morality,

(2) an account of the norms that achieve those objectives, and (3) an account of

normative justification (both pragmatic and coherentist). This defense in terms of

(1) through (3) sets the stage for the next essay in the volume (#11, as abstracted

immediately below).

Essay 11. “From Morality to Common Morality”

Publication Data. This essay is forthcoming in a volume to be entitled, roughly,

Bernard Gert and Applied Philosophy. It is being edited at Dartmouth College as

the present volume goes to press. Published by permission of Jim Moor and

Bernard Gert. �Abstract. For some 19 years, Bernard Gert has criticized my views about moral

philosophy and the principles of biomedical ethics. In this article I focus on the

major issue addressed in his moral theory: the justification of morality—that is,

the justification of the common morality. I concentrate on a body of claims that

he and I both defend about the common morality, and I emphasize our simila-

rities rather than our differences. I orient the discussion around his account and

develop my own account in the process.

My objectives in this paper are threefold: first, to argue that the common

morality is a reasonable basis for both moral theory and practical ethics;

I N T R O D U C T I O N xxi

second, to identify and defend three forms of justification of the common

morality; and third, to show precisely where Gert and I agree and disagree. The

main question I address is, “Which types of justification of the common morality

are needed, and for which types of claims about the common morality are they

suitable?” I present three distinct strategies of justification: (1) normative theore-

tical justification, resting on ethical theory; (2) normative conceptual justification,

resting on conceptual analysis, and (3) empirical justification, resting on empirical

research. Each of these three types justifies a different conclusion about the

common morality. I do not produce a justification of any one of the three

strategies, although I outline the form such justifications would take. That is,

my limited aim is to identify three available types of justification and to identify

the conclusions each type can be expected to reach. In particular, I distinguish

justification of the norms of the commonmorality (Gert’s principal project) from

the justification of claims that the common morality exists (which, as yet, is of

little or no interest to Gert, though is of interest to me).

The first half of the paper is devoted to finding points of agreement in our

theories. In the second half, I am critical of Gert’s lack of attendance to a few

important issues, especially his neglect of empirical claims, including ones that he

himself seems to rely on. On the whole, this essay presents my account of the

methods of justification of common morality, and therefore is not limited to a

criticism of Gert’s views.

Essay 12. “On Eliminating the Distinction Between Applied Ethics and

Ethical Theory”

Publication Data. This essay was published in The Monist 67 (October 1984):

514–531. Copyright � The Monist: An International Quarterly Journal of General

Philosophical Inquiry. Open Court Publishing Co., Chicago, Illinois. Reprinted by

permission.

Abstract. I motivate this 1984 paper by noting that so-called “applied ethics”—a

recently coined term—has become a major growth area in the curricular offerings

in North American philosophy, but that its actual standing in philosophy is

insecure. Many philosophers regard the literature in applied ethics as lightweight

and perhaps philosophically barren. However, I argue that understanding and

teaching the best literature in applied ethics can be as difficult asmasteringmaterial

in more abstract regions of ethical theory, but I also argue that no significant

differences distinguish ethical theory and applied ethics as philosophical activities

or methods. I do not maintain that there are no differences in content. I argue only

that good applied philosophers do what philosophers have always done: They

analyze concepts; submit to critical scrutiny various strategies that are used to

xxii I N T R O D U C T I O N

justify beliefs, policies, and actions; examine hidden presuppositions; and offer

both moral criticism and constructive theories.

Early in the paper I argue for eliminating the distinction between applied ethics

and ethical theory. I turn, in later sections, to methodological considerations,

paying particular attention to the “case method” and the way analytical argument

surrounds it in law, business, andmedicine. I argue that philosophers can profit in

both a scholarly and a pedagogical way from certain uses of the case method. At

various points in the paper I examine some of the work and the claims of Norman

Daniels, Bernard Gert, and Dan Clouser.

Essay 13. “Does Ethical Theory Have a Future in Bioethics?”

Publication Data. This essay was published in the Journal of Law, Medicine &

Ethics 32 (2004): 209–217. � 2004 by the American Society of Law, Medicine &

Ethics.

Abstract. This article assesses whether ethical theory is likely to continue in

upcoming years to play the prominent role it has in the previous 25 years of

published literature and curriculum development in bioethics. What transpired

during these years suggests that the field enjoys a successful and stable marriage to

philosophical ethical theory. However, the marriage became shaky as bioethics

became a more interdisciplinary and practical field. A practical price is paid for

theoretical generality in philosophy, and it is not clear that contemporary

bioethics is willing to pay that price. It is also often unclear whether and, if so,

how theory is to be brought to bear on dilemmatic problems, public policy, moral

controversies, and moral conflict. I envision that the next 25 years could be very

different because of this now troubled marriage. The most philosophical parts of

bioethics seem headed toward a retreat to philosophy departments and philo-

sophy journals, while bioethics continues on its current course toward becoming

a more interdisciplinary and practical field.

One piece of evidence of philosophy’s declining influence is that many individuals

in law, theological ethics, political theory, the social and behavioral sciences, and the

health professions now carefully address mainstream issues of bioethics without

finding ethical theory essential or even particularly useful or insightful. Another is

that philosophers have yet to offer detailed statements of a method for moving from

philosophical theories to the practical commitments of the theories. Although many

moral philosophers are at present actively involved in problems of biomedical ethics

such as clinical and corporate consulting, policy formulation, and committee review,

it is an open question what their role as moral philosophers should be and whether

they can successfully bring ethical theories and methods to bear on problems of

practice.

I N T R O D U C T I O N xxiii

My concerns in this essay are with the types of theory and method that have been

under discussion in bioethics in the last quarter-century. Three interconnected areas

have been prominent: (1) normative moral theories (from utilitarian and Kantian

theories to principlism, casuistry, virtue ethics, feminist ethics, particularism, etc.);

(2) moral and conceptual analyses of basic moral notions (informed consent, the

killing/letting-die distinction, etc.); and (3) methodology (how bioethics

proceeds—e.g., by use of cases, narratives, specified principles, theory application,

reflective equilibrium, legalmethods, etc.). I leave it an open questionwhether (2) or

(3) can be successfully addressed without addressing (1), an unresolved problem in

philosophical ethics. However, I question philosophy’s success in all three areas,

laying emphasis on its weaknesses in connecting theory to practice.

In assessing the contemporary literature and how it needs to change, I confine

attention to three substantive areas of the intersection between bioethics and

ethical theory: cultural relativity and moral universality, moral justification, and

conceptual analysis. In each case I argue that philosophers need to develop

theories and methods more closely attuned to practice. The work of several

philosophers, including Ruth Macklin, Norman Daniels, and Gerald Dworkin,

is examined. In their writings there is a methodological gap between philosophical

theory (andmethod) and practical conclusions. The future of philosophical ethics

in interdisciplinary bioethics may turn on whether such gaps can be closed. If not,

bioethics may justifiably conclude that philosophy is of little practical value for

the field.

Essay 14. “The Failure of Theories of Personhood”




Abstract. This article focuses on the pervasive belief in popular culture, philo-

sophy, religion, law, and science (e.g., in research ethics) that some special

property of persons, such as self-consciousness, rationality, language use, or

dignity, confers a unique moral status. I discuss the distinction between moral

persons and metaphysical persons and also the connection or lack of connection

between the theory of autonomy and the theory of persons. I argue that no set of

cognitive properties alone confers moral standing and that metaphysical person-

hood is not sufficient for either moral personhood or moral standing. Cognitive

theories fail to capture the depth of commitments embedded in using the lan-

guage of “person,” and it is more assumed than demonstrated in cognitive

theories that nonhuman animals lack a relevant form of self-consciousness or

its functional equivalent. Although nonhuman animals are not plausible

xxiv I N T R O D U C T I O N

candidates for moral persons, humans, too, fail to qualify as moral persons if they

lack one or more of the conditions of moral personhood. If moral personhood

were the sole basis of moral rights, then these humans would lack rights for the

same reasons that nonhuman animals are often held to lack rights.

I also argue that the vagueness and the inherently contestable nature of the

concept of person are not likely to be dissipated by philosophical theories of the

nature of persons and that we would be better off if we eliminated the language of

“person” from moral theory altogether and replaced it with more specific

concepts.

Essay 15. “Looking Back and Judging Our Predecessors”




Abstract. This essay is on the problem of retrospective moral judgment. It

considers how moral theory and related methods of assessment should be used

to address the following question: “Can persons and institutions be held respon-

sible for actions taken decades ago, when moral standards, practices, and policies

were strikingly different, or even nonexistent?” The question is whether the

principles and rules that we currently embrace are unfairly retrofitted when we

use them to make judgments about the medical ethics of our predecessors. This

seemingly straightforward question requires making several distinctions and

using different forms of argument to untangle the issue(s). For example, issues

of wrongdoing need to be disengaged from questions of culpability and exculpa-

tion. Also, even if institutions can be found guilty of wrongdoing, it does not

follow that particular individuals in those institutions can be found to be either

wrongdoers or culpable.

To illustrate the problems of theory and method present in these questions,

I consider two morally and politically important examples of how these ques-

tions have arisen in biomedical ethics. Both come from research ethics, and the

two are intimately connected. The first source is a set of moral problems

addressed in the Report of the Advisory Committee on Human Radiation

Experiments that was appointed by President William Clinton to investigate

questionable experiments funded by the U.S. government after World War II.

The second is the work of an Ad Hoc faculty committee at the University of

California, San Francisco (“Report of the UCSF Ad Hoc Fact Finding

Committee on World War II Human Radiation Experiments”) that investi-

gated the ethics of the actions of its own administration and faculty in its

involvement in the human radiation experiments.

I N T R O D U C T I O N xxv

The Advisory Committee identified six basic ethical principles as relevant to its

work and then appropriately argued that persons and institutions can be held

responsible for actions taken even if the standards, practices, and policies at the

time on the use of research subjects were strikingly different than those we call

upon today. I argue that in reaching its conclusions, the Advisory Committee

did not altogether adhere to the language and commitments of its own ethical

framework. In its Final Report, the Advisory Committee emphasizes judgments of

wrongdoing, to the relative neglect of culpability; I argue that the Advisory

Committee properly treats mitigating conditions that are exculpatory, but does

not provide a thoroughgoing assessment of either culpability or exculpation.

It also fails to judge the culpability of particular individuals, though it was

positioned to do so.

I am thus critical of the Advisory Committee’s findings, but I am especially critical

of the more serious deficiencies in the Ad Hoc Committee’s deliberations and

conclusions. The latter group reaches no significant judgments of either wrongdoing

or culpability, but almost certainly should have. A balanced investigation would have

more critically assessed (1) physician wrongdoing, (2) the culpability of specific

agents, and (3) institutional responsibility.

xxvi I N T R O D U C T I O N

Part I

THE BELMONT REPORT AND THE RISE OF

PRINCIPLES

1

THE ORIGINS AND EVOLUTION OF THE

BELMONT REPORT

When, onDecember 22, 1976, I agreed to join the staff of theNationalCommission for

the Protection of Human Subjects of Biomedical and Behavioral Research, my first

and onlymajor assignment was towrite the “Belmont Paper,” as it was then called. At

the time, I had already drafted substantial parts of Principles of Biomedical Ethicswith

James Childress.1 Subsequent to my appointment, the two manuscripts were drafted

simultaneously, often side by side, the one inevitably influencing the other.

I here explain how the “Belmont Paper” evolved into the Belmont Report.2 I will

also correct some common but mistaken speculation about the emergence of

frameworks of principles in research ethics and the connections between Belmont

and Principles.

THE BEGINNINGS OF BELMONT

The idea for the “Belmont Paper” originally grew from a vision of shared moral

principles governing research that emerged during a break-out session at a four-

day retreat held February 13–16, 1976, at the Smithsonian Institution’s Belmont

Conference Center in Maryland.3 Albert Jonsen has reported on the contributions

at this meeting of Stephen Toulmin, Karen Lebacqz, Joe Brady, and others.4

However, this meeting predates my work on the Belmont Report, and I leave it

to Jonsen and the others in attendance to relate the story of the retreat.

A few months after this conference at Belmont, I received two phone calls: the

first fromToulmin, whowas the staff philosopher at the National Commission, and

the second from Michael Yesley, staff director. They asked me to write a paper for

the National Commission on the nature and scope of the notion of justice. Yesley

told me that the commissioners sought help in understanding theories of justice

and their applications to themoral problems of human subject research. I wrote this

paper and assumed that my work for the National Commission was concluded.5

However, shortly after I completed the paper, Toulmin returned to full-time

teaching at the University of Chicago, and Yesley inquired whether I was available

to replace him on the staff. This appointment met some resistance. Two commis-

sioners who later became my close friends—Brady and Donald Seldin—were

initially skeptical of the appointment. Nonetheless, Yesley prevailed, likely with

the help of Chairperson Kenneth Ryan and my colleague Patricia King, and

I joined the National Commission staff.

On my first morning in the office, Yesley told me that he was assigning me the

task of writing the “Belmont Paper.”6 I asked Yesley what the task was. He pointed

out that the National Commission had been charged by Congress to investigate

the ethics of research and to explore basic ethical principles.7Members of the staff

were at work on various topics in research ethics, he reported, but no one was

working on basic principles. He said that an opening round of discussions of the

principles had been held at the Belmont retreat. The National Commission had

delineated a rough schema of three basic ethical principles: respect for persons,

beneficence, and justice. I asked Yesley what these moral notions meant to the

commissioners, to which he responded that he had no well-formed idea and that

it was my job to figure out what the commissioners meant—or, more likely, to

figure out what they should have meant.

So, I found myself with the job of giving shape and substance to something

called the “Belmont Paper,” though at that point I had never heard of Belmont or

the paper. It struck me as an odd title for a publication. Moreover, this document

had never been mentioned during my interview for the job or at any other time

until Yesley gave me the assignment. My immediate sense was that I was the new

kid on the block and had been given the dregs of National Commission work.

I had thought, when I decided to join the National Commission staff, that I would

be working on the ethics of psychosurgery and research involving children, which

were heated and perplexing controversies at the time. I was chagrined to learn that

I was to write something on which no one else was working and that had its

origins in a retreat that I had not attended. Moreover, the mandate to do the work

had its roots in a federal law that I had not seen until that morning.

Yesley proceeded to explain that no one had yet worked seriously on the

sections of the report on principles because no one knew what to do with them.

This moment of honesty was not heartening, but I was not discouraged either,

4 T H E B E L M O N T R E P O R T A N D T H E R I S E O F P R I N C I P L E S

because Childress and I were at that time well into the writing of our book on the

role of basic principles in biomedical ethics. It intrigued me that the two of us had

worked relatively little on research ethics, which was the sole focus of the National

Commission. I saw in my early conversations with Yesley that these two projects,

Principles and Belmont, had many points of intersecting interest and could be

mutually beneficial. And so it would be.

Yesley also gave me some hope by saying that a crude draft of the “Belmont

Paper” already existed, though a twinkle in his eye warned me not to expect too

much. That same morning I read the “Belmont draft.”8 Scarce could a new recruit

have been more dismayed. So little was said about the principles that to call it a

“draft” of principles would be like calling a dictionary entry a scholarly treatise.

Some sections were useful, especially a few handsome pages that had been written

largely by Robert Levine on the subject of “The Boundaries Between Biomedical

and Behavioral Research and Accepted and Routine Practice” (later revised under

the subtitle “Boundaries Between Practice and Research” and made the first

section of the Belmont Report). Apart from Levine’s contribution, however, this

draft of Belmont had almost nothing to say about the principles that were slated to

be its heart.

In the next few weeks, virtually everything in this draft pertaining to principles

would be thrown away either because it contained too little on principles or

because it had too much on peripheral issues. At the time, these peripheral issues

constituted almost the entire document, with the exception of the section written

by Levine, which was neither peripheral nor on principles. The major topics

addressed were the National Commission’s mandate, appropriate review

mechanisms, compensation for injury, national and international regulations

and codes, research design, and other items that did not belong in the Belmont

Report. These topics, being peripheral, were therefore eliminated. Except for

Levine’s section on boundaries, everything in this draft landed on the cutting-

room floor.9

Once the “Belmont draft” was left with nothing in the section on principles,

Yesley suggested that I might find the needed content from the massive compen-

dium on research titled Experimentation with Human Beings, edited by Jay Katz

with the assistance of Alexander Capron and Eleanor Swift Glass.10 Drawn from

sociology, psychology, medicine, and law, this book was at the time the most

thorough collection of materials on research ethics and law. Yesley informed me

that I should endeavor to learn all the information presented in this book, but

after days of poring over this wonderful resource, I found that it offered virtually

nothing on principles suitable for an analytical discussion of research ethics. The

various codes and statements by professional associations found in this book had

occasional connections with my task and with the National Commission’s objec-

tives, but only distant ones.11

T H E O R I G I N S A N D E V O L U T I O N O F T H E B E L M O N T R E P O R T 5

THE HISTORICAL ORIGINS OF THE PRINCIPLES

OF THE BELMONT REPORT

Fortunately, Childress and I had gathered a useful collection of materials on

principles and theories, largely in the writings of philosophers. I had been

influenced in my thinking about principles by the writings of W. D. Ross and

William Frankena. My training led me to turn to these and other philosophical

treatments, which had already proved helpful in my work on Principles.

However, it would be misleading to suggest that the principles featured in the

Belmont Report derived from the writings of philosophers. Their grounding is

ultimately in what I would eventually call, following Alan Donagan, “the common

morality.” The Belmont Report makes reference to “our cultural tradition” as the

basis of its principles, and it is clear that these principles derive from the common

morality rather than a particular philosophical work or tradition. However, what

Belmont means by our “tradition” is unclear, and I believe the import of the

Belmont principles cannot be tied to a particular tradition, but rather to a

conviction that there is a universally valid point of view. I believe, and I think

that the commissioners believed, that these principles are norms shared by all

morally informed and committed persons.

The commissioners almost certainly believed that these principles are already

embedded in public morality and are presupposed in the formulation of public and

institutional policies. The principles do not deviate from what every morally sensi-

tive personknows tobe right, based onhis orher ownmoral training and experience.

That is, every morally sensitive person believes that a moral way of life requires that

we respect persons and take into account their well-being in our actions. Belmont’s

principles are so woven into the fabric of morality that no responsible research

investigator could conduct research without reference to them.12

THE RELATIONSHIP BETWEEN THE BELMONT PRINCIPLES AND THE

PRINCIPLES IN PRINCIPLES OF BIOMEDICAL ETHICS

Many have supposed that the Belmont Report provided the starting point and the

abstract framework for Principles of Biomedical Ethics.13 They have wrongly

assumed that Belmont preceded and grounded Principles.14 The two works were

written simultaneously, the one influencing the other.15 There was reciprocity in

the drafting, and influence ran bilaterally. I was often simultaneously drafting

material on the same principle or topic both for the National Commission and for

my colleague Childress, while he was at the same time writing material for me to

inspect. I would routinely write parts of the Belmont Report during the day at the

National Commission headquarters on Westbard Avenue in Bethesda, then go to


my office at Georgetown University in the evening and draft parts of chapters for

Childress to review. Despite their entirely independent origins, these projects

grew up and matured together.

Once I grasped the moral vision the National Commission initiated at Belmont, I

could see thatChildress and I hadmajor substantive disagreementswith theNational

Commission. The names of the principles articulated in the Belmont Report bear

notable similarities to some of the names Childress and I were using and continued

to use, but the two schemas of principles are far from constituting a uniform name,

number, or conception. Indeed, the two frameworks are inconsistent.

I regarded the National Commission as confused in the way it analyzed the

principle of respect for persons. It seemed to blend a principle of respect for

autonomy and a principle of beneficence that required protection of, and avoidance

of the causation of harm to, incompetent persons. Childress and I both thought at

the time that we should make our analysis turn on sharp distinctions between the

principle of beneficence, the principle of nonmaleficence, and the principle of

respect for autonomy, though the National Commission did not make such sharp

distinctions. This matter was connected to an additional problem: The National

Commission may have had a fundamentally utilitarian vision of social beneficence

without adequate internal controls in its framework toprotect subjects against abuse

when there is the promise of major benefit for other sick or disabled persons.

The differences between the philosophy in Principles and the National

Commission’s views in Belmont have occasionally been the subject of published

commentary.16 Some commentators correctly see that we developed substantially

different moral visions and that neither approach was erected on the foundations

of the other. By early 1977, I had come to the view that the National Commission,

especially in the person of its chair, Kenneth Ryan, was sufficiently rigid in its

vision of the principles that there was no way to substantively alter the conception,

although I thought that all three principles were either defective or underanalyzed.

From this point forward, I attempted to analyze principles for the National

Commission exclusively as the commissioners envisioned them. Principles of

Biomedical Ethics became the sole work expressing my deepest philosophical

convictions about principles.

THE DRAFTING AND REDRAFTING OF BELMONT

While Yesley gave me free rein in the drafting and redrafting of Belmont, the drafts

were always subject to revisions and improvements made by the commissioners and

staff members.17 All members of the staff did their best to formulate ideas that were

responsive to changes suggestedby the commissioners.Commissioner Seldin encour-

aged me with as much vigor as he could muster, which was and remains today


considerable, to make my drafts as philosophical as possible. Seldin wanted some

Mill here, some Kant there, and the signature of philosophical argument sprinkled

throughout the document. I tried this style, but other commissioners wanted a

streamlined document and minimalist statement relatively free of the trappings of

philosophy. Seldin, Yesley, and I ultimately relented,18 and bolder philosophical

defenses of the principles were gradually stripped from the body of Belmont.

Public deliberations in National Commission meetings were a staple source of

ideas, but a few commissioners spoke privately to me or to Yesley about desired

changes, and a few commissioners proposed changes to Assistant Staff Director

Barbara Mishkin, who then passed them on to me. Most of these suggestions were

accepted, and a serious attempt was made to implement them. In this respect, the

writing of this document was a joint product of commissioner–staff interactions.

However, most of the revisions made by the commissioners, other than through

their comments in public deliberations, concerned small matters, and the commis-

sioners were rarely involved in tendering written changes or detailed suggestions.

An exception to this generalization is found in a meeting on Belmont involving

a few members of the staff and a few commissioners that occurred during

September 1977 in the belvedere (or rooftop study) of Jonsen’s home in San

Francisco. The small group in attendance attempted to revise the “Belmont

Paper” for presentation at the next meetings during which the commissioners

were scheduled to debate it. As Jonsen reports in The Birth of Bioethics and

elsewhere, the purpose of this meeting was to revisit previous drafts and delibera-

tions of the commissioners.19

The history of drafting and redrafting that I have outlined may suggest that the

document grew in size over time, but the reverse is true. The document grew

quickly in the early drafting and then was contracted over time. I wrote much

more for the National Commission about respect for persons, beneficence, and

justice than eventually found its way into the Belmont Report. When various

materials I had written were eliminated from Belmont, I would scoop up the

reject piles and fashion them for Principles of Biomedical Ethics. Several late-

written chunks of this book, in its parts on research ethics, were fashioned from

the more philosophical but abjured parts of what I wrote for the National

Commission that never found their way into the final draft of Belmont.

EXPLICIT AND IMPLICIT IDEAS ABOUT AN APPLIED RESEARCH ETHICS

Michael Yesley deserves credit for one key organizing conception in this report.

He and I were in almost daily discussion about the “Belmont Paper.” We spent

many hours discussing the best way to develop the principles to express what the

commissioners wanted to say, and even how to sneak in certain lines of thought


that the commissioners might not notice. One late afternoon we were discussing

the overall enterprise. We discussed each principle, whether the principles were

truly independent, and how the principles related to the topics in research ethics

under consideration by the National Commission. Yesley said, as a way of

summarizing our reflections, “What these principles come to is really quite

simple for our purposes: Respect for persons applies to informed consent, ben-

eficence applies to risk-benefit assessment, and justice applies to the selection of

subjects.” Yesley had articulated the following abstract schema:

This schema may seem trifling today, and it was already nascent in preexisting

drafts of the report and in the National Commission’s deliberations. But no one at

the time had articulated these ideas in precisely this way, and Yesley’s summary

was immensely helpful in peering through countless hours of discussion to see the

underlying structure and commitment at work in the principles destined to be the

backbone of Belmont. Yesley had captured what would soon become the major

portion of the table of contents of the Belmont Report, as well as the rationale of its

organization. I then attempted to draft the document so that the basic principles

could be applied to develop guidelines in specific areas and could also serve as

justification for guidelines.

In light of this schema, a general strategy emerged for handling problems of

research ethics, namely, that each principle made moral demands in a specific

domain of responsibility for research. For example, the principle of respect for

persons demands informed and voluntary consent. Under this conception, the

purpose of consent provisions is not protection from risk, as many earlier federal

policies seemed to imply, but the protection of autonomy and personal dignity,

including the personal dignity of incompetent persons incapable of acting auton-

omously, for whose involvement a duly authorized third party must consent.

I wrote the sections on principles in the Belmont Report based on this model of

each principle applying to a zone of moral concern. In this drafting, the focus of

the document shifted to include not only abstract principles and their analysis,

but also a moral view that is considerably more concrete andmeaningful for those

engaged in the practice of research. Explication of the value being advanced was

heavily influenced by the context of biomedicine, and rather less influenced by

contexts of the social and behavioral sciences. Belmont, in this way, moved toward

an applied, professional morality of research ethics.

Principle of Applies to Guidelines for

Respect for persons Informed consent

Beneficence Risk–benefit assessment

Justice Selection of subjects


Although Belmont takes this modest step in the direction of an applied research

ethics, there was never any ambition or attempt to make this document specific

and practical. This objective was to be accomplished by the other volumes the

National Commission issued. Belmont was meant to be, and should be remem-

bered as, a moral framework for research ethics. Commissioners and staff were

always aware that this framework is too indeterminate by itself to decide practice

or policy or to resolve moral conflicts. The process of making the general

principles in Belmont sufficiently concrete is a progressive process of reducing

the indeterminateness and abstractness of the general principles to give them

increased action-guiding capacity.20 Belmont looks to educational institutions,

professional associations, government agencies, institutional review boards, and

the like to provide the particulars of research ethics.

PRINCIPLISM, CASUISTRY, OR BOTH?

Another misunderstanding of both Principles and Belmont has emerged in recent

treatments of casuistry. Some commentators have misinterpreted the National

Commission’s work by overreading it as fundamentally a casuistry, and other

commentators have misinterpreted Principles of Biomedical Ethics by under-

reading it as competitive with, and possibly undermined by, casuistry. However,

a better interpretation is available: The National Commission demonstrably used

casuistical reasoning in its deliberations, but its commitment to basic principles

was never challenged throughout its years of deliberation. Similarly, Childress and

I have from our first edition used casuistical reasoning in the treatment of cases,

and we do not believe that there is any inconsistency between this form of

reasoning and a commitment to principles. In the histories of both the National

Commission and Principles of Biomedical Ethics, it is difficult to understand how

the two commitments could be seen as other than complementary.

The final editing of the Belmont Report was done by three people in a small

classroom at the NIH.21 Al Jonsen, Stephen Toulmin, and I were given this assign-

ment by the National Commission. Some who have followed the later writings of

Jonsen and Toulmin on casuistry may be surprised to learn that throughout the

National Commission’s deliberations, as well as in this final drafting, Jonsen and

Toulmin contributed to the clarification of the principles in the report. There was

never an objection by either that a strategy of using principles should be other than

central to the National Commission’s statement of its ethical framework. Jonsen has

often repeated his support for these principles.

However, Jonsen has also said that commissioners stated their views as

principlists, but deliberated as casuists. He is suggesting that the National

Commission’s deliberations constituted a casuistry of reasoning about historical


and contemporary cases,22 despite the commissioners’ commitment to and frequent

reference to the Belmont principles. Jonsen and Toulmin once explicated this under-

standingof theNationalCommission’swork in termsof the individual commissioners

coming to agreement not on matters of moral theory and universal principles, but

rather onhow tomake judgments about specific kinds of cases. Agreement, they argue,

could never have been derived from the principles.23

This interpretation gives insight into theNational Commission, but it needs careful

qualification to avoid misunderstanding. Casuistical reasoning more so than moral

theory or universal abstraction often did function to forge agreement duringNational

Commission deliberations. The commissioners appealed to particular cases and

families of cases, and consensuswas reached through agreement on cases and general-

ization from cases when agreement on an underlying theoretical rationale would have

been impossible.24 Commissioners would never have been able to agree on a single

ethical theory, nor did they even attempt to buttress the Belmont principles with a

theory. Jonsen andToulmin’s treatment of theNationalCommission is, in this regard,

entirely reasonable, and a similar line of argument can be taken to explicate the

methods of reasoning at work in other bioethics commissions.25

Nonetheless, thismethodological appraisal is consistent with a firm commitment

to moral principles; the commissioners, including Jonsen, were emphatic in their

support of and appeals to the general moral principles delineated in the Belmont

Report.26 The transcripts of the National Commission’s deliberations show a con-

stant movement from principle to case and from case to principle. Principles

supported arguments about how to handle a case, and precedent cases supported

the importance of commitment to principles. Cases or examples favorable to one

point of view were brought forward, and counterexamples then advanced.

Principles were invoked to justify the choice and use of both examples and counter-

examples. On many occasions an argument was offered that a case judgment was

irrelevant or immoral in light of the commitments of a principle.27 The National

Commission’s deliberations and conclusions are best understood in terms of

reasoning in which principles are interpreted and specified by the force of examples

and counterexamples that emerge from experience with cases.

It is doubtful that Jonsen ever intended to deny this understanding of principles

and their roles, despite the widely held view that casuistry dispenses with principles.

Jonsen has said that “casuistic analysis does not deny the relevance of principle and

theory,”28 and, in an insightful statement in his later work, he has written that:

when maxims such as “Do no harm,” or “Informed consent is obligatory,” are

invoked, they represent, as it were, cut-down versions of the major principles

relevant to the topic, such as beneficence and autonomy, cut down to fit the

nature of the topic and the kinds of circumstances that pertain to it.29


Jonsen goes on to point out that casuistry is “complementary to principles” and

that “casuistry is not an alternative to principles: No sound casuistry can dispense

with principles.”30

Casuists and those who support frameworks of principles like those in Belmont

and Principles of Biomedical Ethics should be able to agree that when they reflect

on cases and policies, they rarely have in hand either principles that were for-

mulated without reference to experience with cases or paradigm cases lacking a

prior commitment to general norms. Only a false dilemma makes us choose

between the National Commission as principlist or casuist. It was both.

Notes

1. A contract for the book was issued by Oxford University Press on August 19, 1976. The

manuscript was completed in late 1977; galleys arrived in October 1978, bearing the 1979

copyright date.

2. See also Appendices I and II to the Belmont Report. The Belmont Reportwas completed

in late 1977 and published on September 30, 1978: The Belmont Report: Ethical Guidelines for

the Protection of Human Subjects of Research (Washington, DC: DHEW Publication OS

78-0012). It first appeared in the Federal Register on April 18, 1979.

The National Commission for the Protection of Human Subjects of Biomedical and

Behavioral Research was established July 12, 1974, under the National Research Act, Public

Law 93-348, Title II. The first meeting was held December 3–4, 1974. The 43rd and final

meeting was on September 8, 1978.

3. See the archives of the National Commission, 15th meeting, February 13–16, 1976, vol.

15A—a volume prepared for the Belmont meeting. This meeting book contains a “staff

summary” on the subject of “ethical principles” as well as expert papers prepared by Kurt

Baier, Alasdair Maclntyre, James Childress, H. Tristram Engelhardt, Alvan Feinstein, and

LeRoy Walters. The papers by Engelhardt andWalters most closely approximate the moral

considerations ultimately treated in the “Belmont Paper,” but neither quite matches the

National Commission’s three principles. Walters, however, comes very close to a formula-

tion of the concerns in practical ethics to which the National Commission applies its

principles. All meeting books are housed in the archives of the Kennedy Institute of Ethics

Library, Georgetown University.

4. “On the Origins and Future of the Belmont Report,” in Belmont Revisited: Ethical

Principles for Research, ed. James F. Childress, Eric M. Meslin, and Harold T. Shapiro

(Washington, DC: Georgetown University Press, 2005), 3–11.

5. The paper was published as “Distributive Justice and Morally Relevant Differences,”

in Appendix I to the Belmont Report, 6.1–6.20. This paper was distributed at the 22nd

meeting of the National Commission, held in September 1976, seven months after the

retreat at the Belmont Conference House.

6. My first day in attendance at a National Commission meeting was the Saturday

meeting of the National Commission on January 8, 1977.


7. The National Research Act, P.L. 93-348, July 12, 1974. Congress charged the National

Commission with recommending regulations to the Department of Health, Education, and

Welfare (DHEW) to protect the rights of research subjects and developing ethical princi-

ples to govern the conduct of research. In this respect, the Belmont Report was at the core of

the tasks the National Commission had been assigned by Congress. DHEW’s conversion of

its grants administration policies governing the conduct of research involving human

subjects into formal regulations applicable to the entire department was relevant to the

creation of the National Commission. In the U.S. Senate, Senator Edward Kennedy, with

Jacob Javits’s support, was calling for a permanent, regulatory commission independent of

the National Institutes of Health (NIH) to protect the welfare and rights of human subjects.

Paul Rogers in the House supported the NIH in advocating that the commission be

advisory only. Kennedy agreed to yield to Rogers if DHEW published satisfactory regula-

tions. This compromise was accepted. Regulations were published on May 30, 1974; then,

on July 12, 1974, P.L. 93-348 was modified to authorize the National Commission as an

advisory body. Charles McCarthy helped me understand this history. For a useful framing

of the more general regulatory history, see Joseph V. Brady and Albert R. Jonsen (two

commissioners), “The Evolution of Regulatory Influences on Research with Human

Subjects,” in Human Subjects Research, ed. Robert Greenwald, Mary Kay Ryan, and

James E. Mulvihill, (New York: Plenum Press, 1982), 3–18.

8. This draft had a history beginning with the 16th meeting (March 12–14, 1976), which

contained a draft datedMarch 1, 1976, and titled “Identification of Basic Ethical Principles.”

This document summarized the relevant historical background and set forth the three

“underlying ethical principles” that came to form the National Commission’s framework.

Each principle was discussed in a single paragraph. This document was slightly recast in a

draft of June 3, 1976 (prepared for the ninth meeting, June 11–13, 1976), in which the

discussion of principles was shortened to little more than one page devoted to all three

principles. Surprisingly, in the summary statement (p. 9), “respect for persons” is presented

as the principle of “autonomy.” No further draft was presented to the National

Commission until ten months later, at the 29th meeting (April 8–9, 1977). I began work

on the document in January 1977.

Transcripts of the National Commission’s meetings are also available in the archives of

the Kennedy Institute of Ethics at Georgetown University. See National Commission for

the Protection of Human Subjects of Biomedical and Behavioral Research. Archived

Materials 1974–78, General Category: “Transcript of the Meeting Proceedings” (for discus-

sion of the “Belmont Paper” at the following meetings: February 11–13, 1977; July 8–9, 1977;

April 14–15, 1978; and June 9–10, 1978).

9. Cf. the radical differences between the draft available at the 19th meeting (June 1–13,

1976) and the draft at the 29th meeting (April 8–9, 1977). The drafts show that the critical

period that gave shape to the Belmont principles was the period between January and April

1977. Less dramatic improvements were made between April 1977 and eventual publication

more than a year later.

10. Jay Katz, with the assistance of Alexander Capron and Eleanor Glass, eds.,

Experimentation with Human Beings (New York: Russell Sage Foundation, 1972).


11. The first and only footnote in the Belmont Report is a reference to this background

reading. Typical materials that I examined during this period include United States v. Karl

Brandt, Trials of War Criminals Before the Nuremberg Military Tribunals Under Control

Council Law No. 10, 1948–49, Military Tribunal I (Washington, DC: U.S. Government

Printing Office, 1948–1949), vols. 1 and 2, reproduced in part in Katz, Experimentation with

Human Beings, 292–306; American Medical Association, House of Delegates, Judicial

Council, “Supplementary Report of the Judicial Council,” Journal of the American

Medical Association 132 (1946): 90; World Health Organization, 18th World Medical

Assembly, Helsinki, Finland, “Declaration of Helsinki: Recommendations Guiding

Medical Doctors in Biomedical Research Involving Human Subjects,” New England

Journal of Medicine 271 (1964): 473, reprinted in Katz, Experimentation with Human

Beings, 312–313. Less helpful than I had hoped was Henry Beecher, “Some Guiding

Principles for Clinical Investigation,” Journal of the American Medical Association 195

(1966): 1135–1136. For behavioral research, I started with Stuart E. Golann, “Ethical

Standards for Psychology: Development and Revisions, 1938-1968,” Annals of the New

York Academy of Sciences 169 (1970): 398–405, and American Psychological Association,

Inc., Ethical Principles in the Conduct of Research with Human Participants (Washington,

DC: APA, 1973).

12. For a clearer presentation of this viewpoint in a later document by a government-

initiated commission, see several chapters in Advisory Committee on Human Radiation

Experiments (ACHRE), Final Report of the Advisory Committee on Human Radiation

Experiments (New York: Oxford University Press, 1996).

13. See, e.g., Eric Meslin et al., “Principlism and the Ethical Appraisal of Clinical Trials,”

Bioethics 9 (1995): 399–403; Bernard Gert, Charles M. Culver, and K. Danner Clouser,

Bioethics: A Return to Fundamentals (New York: Oxford University Press, 1997), 72–74; and

Jonathan D. Moreno, Deciding Together: Bioethics and Consensus (New York: Oxford

University Press, 1995), 76–78. Meslin et al. say that “Beauchamp and Childress’s

Principles of Biomedical Ethics . . . is themost rigorous presentation of the principles initially

described in the Belmont Report” (p. 403). Gert et al. see Principles as having “emerged from

the work of the National Commission” (p. 71). Moreno presumes that Beauchamp and

Childress “brought the three [Belmont] principles into bioethical analysis more generally.”

The thesis that the idea of an abstract framework of principles for bioethics originated with

the National Commission has been sufficiently prevalent that authors and lecturers have

occasionally cited the principles as Childress and I have named and articulated them and

then felt comfortable in attributing these same principles to the National Commission.

14. The draft of Belmont that appeared in typescript for the National Commission

meeting of December 2, 1977 (37th meeting) shows several similarities to various passages

in the first edition of Principles of Biomedical Ethics. Childress and I had completed our

manuscript by this date, but Belmont would be taken through five more drafts presented to

the commissioners, the last being presented at the final (43rd) meeting (September 8, 1978).

15. Prior to my involvement with the National Commission, and prior to the Belmont

retreat, Childress and I had lectured and written about principles of biomedical ethics. In

early 1976, coincidentally at about the same time of the Belmont retreat, Childress and I


wrote a programmatic idea for the book, based on our lectures, which we submitted to

Oxford University Press.We had already developed a general conception of what later came

to be called by some commentators “midlevel principles.”

For more on the nature, history, and defensibility of a commitment to such midlevel

principles in bioethics, see Gert, Culver, and Clouser, Bioethics: A Return to Fundamentals,

72ff; Tom Beauchamp and David DeGrazia, “Principlism,” in Bioethics: A Philosophical

Overview, vol. 1 of Handbook of the Philosophy of Medicine, ed. George Khusfh (Dordrecht,

Netherlands: Kluwer, 2002); James Childress, “Ethical Theories, Principles, and Casuistry

in Bioethics: An Interpretation and Defense of Principlism,” in Religious Methods and

Resources in Bioethics, ed. Paul F. Camenisch (Boston: Kluwer, 1994), 181–201; Beauchamp,

“The Four Principles Approach toMedical Ethics,” in Principles of Health Care Ethics, ed. R.

Gillon, (London: John Wiley & Sons, 1994), 3–12; Beauchamp, “The Role of Principles in

Practical Ethics,” in Philosophical Perspectives on Bioethics, ed. L. W. Sumner and J. Boyle

(Toronto: University of Toronto Press, 1996); and Earl Winkler, “Moral Philosophy and

Bioethics: Contextualism versus the Paradigm Theory,” in Philosophical Perspectives on

Bioethics, ed. L. W. Sumner and J. Boyle (Toronto: University of Toronto Press, 1996).

16. Particularly insightful is Ernest Marshall, “Does theMoral Philosophy of the Belmont

Report Rest on aMistake?” IRB: A Review of Human Subjects Research 8 (1986): 5–6. See also

John Fletcher, “Abortion Politics, Science, and Research Ethics: Take Down the Wall of

Separation,” Journal of Contemporary Health Law and Policy 8 (1992): 95–121, esp. sect. 4,

“Resources in Research Ethics: Adequacy of the Belmont Report.”

17. Albert Jonsen, a commissioner, reports in The Birth of Bioethics (New York: Oxford

University Press, 1998) that I was “working with [Stephen] Toulmin on subsequent drafts”

of the Belmont Report in 1977. Although I sat next to Stephen in National Commission

meetings and conversed with him about many subjects during meetings of the National

Commission throughout 1977, we never jointly drafted, worked on, or discussed Belmont

until it was in near-final form and already approved by the commissioners. Stephen had

been assigned to a project on recombinant DNA and did not participate in Belmont drafts

after I came to the National Commission.

18. Yesley was my constant critic, more so than anyone else. Seldin was my constant

counsel, forever exhorting me to make the document more philosophically credible.

Patricia King taught me more about the National Commission and its commissioners

than anyone else. It was she who helped me understand why a really philosophical

document was not the most desirable result.

19. Jonsen reports in The Birth of Bioethics (p. 103) that “the date is uncertain” of this

meeting at his home (and see his “On the Origins and Future of the Belmont Report,” p. 5).

In fact, the date is the afternoon of September 21 through September 23, 1977 (which

includes the travel period). Jonsen correctly remarks that one purpose of this meeting was

to revisit “the February 1977 deliberations” of the commissioners, but he incorrectly reports

that the meeting was called “to revise the June 1976 draft.” Except for a section on

boundaries written by Robert Levine, the June 1976 draft had been so heavily recast that

Belmont was by September a completely different document. Seven months of continual

redrafting of Belmont had occurred, from February to September 1977, prior to the meeting


at his home. The National Commission did not discuss the draft reports at its meetings

during those seven months. (The last discussion occurred in February). However, “staff

drafts” of the “Belmont Paper” were distributed at two meetings during this period,

namely, the 29th meeting (April 8–9, 1977) and the 30th meeting (May 13–14, 1977). All

drafts are now housed in the archives of the Kennedy Institute of Ethics Library.

20. See Henry S. Richardson, “Specifying Norms as a Way to Resolve Concrete Ethical

Problems,” Philosophy and Public Affairs 19 (1990): 279–310; Richardson, “Specifying,

Balancing, and Interpreting Bioethical Principles,” Journal of Medicine and Philosophy 25

(2000): 285–307, an updated version of which appears in Belmont Revisited: Ethical

Principles for Research, ed. Childress, Meslin, and Shapiro, 205–227. See also Tom

Beauchamp and James Childress, Principles of Biomedical Ethics, 5th ed. (New York:

Oxford University Press, 2001), 15–19; David DeGrazia, “Moving Forward in Bioethical

Theory: Theories, Cases, and Specified Principlism,” Journal of Medicine and Philosophy 17

(1992): 511–539; DeGrazia and Beauchamp, “Philosophical Foundations and Philosophical

Methods,” in Method in Medical Ethics, ed. Jeremy Sugarman and Daniel P. Sulmasy

(Washington, DC: Georgetown University Press, 2001), esp. 31–46.

21. I have diary notes that our editing meetings occurred May 31–June 1, 1978, just prior

to the National Commission’s 42nd and penultimate meeting, on June 9–10, 1978. Thus, the

wording at the 42nd meeting was the final wording unless a commissioner raised an

objection at that meeting or the next.

22. As used here, casuistry implies that some forms of moral reasoning and judgment

neither appeal to nor rely upon principles and rules, but rather involve appeals to the

grounding of moral judgment in narratives, paradigm cases, analogies, models, classifica-

tion schemes, and even immediate intuition and discerning insight. Each change in the

circumstances changes the case. The casuistic method begins with cases whose moral

features and conclusions have already been decided and then compares the salient features

in the paradigm cases with the features of cases that require a decision. See Albert Jonsen

and Stephen Toulmin, Abuse of Casuistry (Berkeley: University of California Press, 1988),

251–254, 296–299; Jonsen, “On the Origins and Future of the Belmont Report,” in Belmont

Revisited, 3–11; Jonsen, “Casuistry as Methodology in Clinical Ethics,” Theoretical Medicine

12 (1991): 299–302; John Arras, “Principles and Particularity: The Role of Cases in

Bioethics,” Indiana Law Journal 69 (1994): 983–1014. See also Toulmin, “The Tyranny of

Principles,” Hastings Center Report 11 (1981): 31–39.

23. Jonsen and Toulmin, Abuse of Casuistry, 16–19.

24. A few years ago, I reviewed all the National Commission transcripts pertaining to

Belmont, primarily to study the National Commission’s method of treating issues in

research ethics. I found that Jonsen and Toulmin had occasionally mentioned casuistry,

but they clearly understood the National Commission’s casuistry as consistent with its

invocation of moral principles. For additional discussion, see Stephen Toulmin, “The

National Commission on Human Experimentation: Procedures and Outcomes,” in

Scientific Controversies: Case Studies in the Resolution and Closure of Disputes in Science

and Technology, ed. H. T. Engelhardt Jr. and A. Caplan (New York: Cambridge University

Press, 1987), 599–613, and Jonsen, “Casuistry as Methodology in Clinical Ethics.”


25. See, e.g., Alexander Capron, “Looking Back at the President’s Commission,”

Hastings Center Report 13, no. 5 (1983): 8–9.

26. See Jonsen’s own summation to this effect in “Casuistry,” Methods of Bioethics, ed.

Sugarman and Sulmasy, 112–113, and his Introduction to a reprinting of the Belmont Report

in Source Book in Bioethics, ed. Albert Jonsen, Robert M. Veatch, and LeRoy Walters

(Washington, DC: Georgetown University Press, 1998).

27. Albert Jonsen, “Case Analysis in Clinical Ethics,” Journal of Clinical Ethics 1 (1990): 65.

28. Jonsen, “Case Analysis in Clinical Ethics”: 65.

29. Jonsen, “Casuistry: An Alternative or Complement to Principles?” Kennedy Institute

of Ethics Journal 5 (1995): 237–251.

30. Jonsen, “Casuistry: An Alternative or Complement to Principles?” 244–249. See also

Jonsen, “The Weight and Weighing of Ethical Principles,” in The Ethics of Research

Involving Human Subjects: Facing the 21st Century, ed. Harold Y. Vanderpool (Frederick,

MD: University Publishing Group, 1996), 59–82.


2

CODES, DECLARATIONS, AND OTHER ETHICAL

GUIDANCE FOR HUMAN SUBJECTS RESEARCH:

THE BELMONT REPORT

The Belmont Report is a short document onmoral principles that was published in

1978 by the National Commission for the Protection of Human Subjects

of Biomedical and Behavioral Research (hereafter the National Commission).

Since that time it has provided a basic framework for analyzing ethical issues that

arise during medical research in the United States as well as in other countries.

HISTORY

TheNational Commission was established in 1974 by the U.S. Congress with a charge

to identify ethical principles and develop guidelines to govern the conduct of research

involving humans. It was hoped the guidelines would ensure that the basic ethical

principles would become embedded in the U.S. research oversight system, so that

meaningful protection was afforded to research participants. Anothermandated goal

was to distinguish the boundaries between the accepted and routine practice of

medicine on the one hand, and biomedical and behavioral research on the other.

The National Commission held its first meeting on December 3–4, 1974, and its

43rd and final meeting on September 8, 1978.1 It completed a draft of the Belmont

Report in late 1977, and issued it in final published form on September 30, 1978.

The report was then republished in the Federal Register on April 18, 1979—a date

now commonly cited as the original date of publication.

The National Commission also published 16 other reports and appendix volumes,

most focused on ethical issues in research involving vulnerable populations. Its more

than 100 recommendations for reform went directly to the Secretary of the

Department of Health, Education and Welfare (DHEW)—later remodeled and

made the Department of Health and Human Services (DHHS). Many of

the National Commission’s recommendations were eventually codified as federal

regulations.2 The Belmont Report itself was not written in the style of federal regula-

tions and was never so codified. It was the National Commission’s statement of a

general, principledmoral framework. The foundation of this “analytical framework,”

as it is called in the report, was a collection of moral principles appropriate for

research that first emerged during discussions at a retreat the National Commission

held on February 13–16, 1976, at the Smithsonian Institution’s Belmont Conference

Center in Elkridge, Maryland. There had been no draft or planning for this frame-

work prior to the retreat. This conference center’s name was then appropriated, and

the report was published under the full title of The Belmont Report: Ethical Principles

and Guidelines for the Protection of Human Subjects of Research.

The National Commission came into existence in the aftermath of public

outrage and congressional uncertainty over the Tuskegee syphilis experiments

and other questionable uses of humans in research. The socioeconomic depriva-

tion of the African American men who were enrolled in the Tuskegee experiments

made them vulnerable to overt and unjustifiable forms of manipulation at the

hands of health professionals, as had been widely reported in news media and

widely circulated in the report of an advisory panel to the DHEW in 1973.3 Other

reports of the abuse of fetuses, prisoners, children, and “the institutionalized

mentally infirm” appeared in the news media.

The law that created the National Commission specified that no more than 5

of the Commission’s 11members could be research investigators (see Table 1). This

stipulation testified to congressional determination at the time that research activ-

ities of the biomedical and behavioral sciences be brought under the critical eye,

and possibly the control, of persons outside of the sciences. At that time, the

research system generally placed responsibility for the protection of humans in

research on the shoulders of individual investigators. That is, federal policies relied

on the discretion and good judgment of investigators to determine the conditions

under which research should be conducted. Federal involvement and review com-

mittees were then in the formative stages. They were destined to undergo rapid

change toward protectionism under the guidance of the National Commission.

Carol Levine offers the following sobering and accurate statement of the

context in which the National Commission deliberated:

The Belmont Report . . . reflected the history of the 30 years immediately

preceding it. This emphasis is understandable, given the signal event in the

C O D E S , D E C L A R A T I O N S , A N D O T H E R E T H I C A L G U I D A N C E 19

modern history of clinical-research ethics [Nazi experimentation]. American

public opinion was shaped by the revelations of unethical experiments such

as the Willowbrook hepatitis B studies, . . . the Jewish Chronic Disease

Hospital studies, . . . and, especially, the Tuskegee Syphilis Study . . . . Our

TABLE 2.1 Members of the National Commission for the Protection of Human Subjects

of Biomedical and Behavioral Research

� Kenneth John Ryan, M.D., Chairman, Chief of Staff, Boston Hospital for Women� Joseph V. Brady, Ph.D., Professor of Behavioral Biology, Johns Hopkins University� Robert E. Cooke, M.D., President, Medical College of Pennsylvania� Dorothy I. Height, President, National Council of Negro Women, Inc.� Albert R. Jonsen, Ph.D., Associate Professor of Bioethics, University of California at San

Francisco� Patricia King, J.D., Associate Professor of Law, Georgetown University Law Center� Karen Lebacqz, Ph.D., Associate Professor of Christian Ethics, Pacific School of Religion� David W. Louisell, J.D., Professor of Law, University of California at Berkeley� Donald W. Seldin, M.D., Professor and Chairman, Department of Internal Medicine,

Southwestern Medical School, University of Texas� Eliot Stellar, Ph.D., Provost of the University and Professor of Physiological Psychology,

University of Pennsylvania� Robert H. Turtle, LL.B., Attorney, VomBaur, Coburn, Simmons and Turtle, Washington, D.C.

FIGURE 2.1 Commissioners and Staff of the National Commission for the Protection ofHuman Subjects of Biomedical and Behavioral Research, 1977.


basic approach to the ethical conduct of research and approval of

investigational drugs was born in scandal and reared in protectionism.

Perceived as vulnerable, either because of their membership in groups

lacking social power or because of personal characteristics suggesting a

lack of autonomy, individuals were the primary focus of this concern.4

CONTENT AND CORE STRENGTHS

The Belmont Report is best known for its framework of basic moral principles,

which are still today referred to as the “Belmont principles.” The National

Commission identified three general principles as underlying the conduct of

research: respect for persons, beneficence, and justice. The key organizing

conception underlying the National Commission’s presentation of these prin-

ciples and their use was the following: Respect for persons applies to informed

consent; beneficence applies to risk–benefit assessment; and justice applies to

the selection of research participants. The following abstract schema represents

this conception:

In this way, each moral principle makes moral demands in a specific domain of

responsibility for research. This conception of the connection between abstract

moral principles and applied bioethics has been enduring. Many engaged in

research ethics carry this general conception with them today.

The principle of respect for persons demands that the choices of autonomous

persons not be overridden or otherwise disrespected and that persons who are

not adequately autonomous be protected by the consent of an authorized third

party who is likely to appreciate their circumstances and will look after their best

interests. This principle in effect requires valid permission before investigators

can proceed with research. To achieve this goal, the principle insists on the

individual’s informed consent, analyzed in terms of the conditions of informa-

tion disclosure, comprehension, and voluntariness. The National Commission

proposed “the reasonable volunteer” as an appropriate standard for judging the

Principle of Applies to

Respect for persons Informed consent

Beneficence Risk–benefit assessment

Justice Selection of research subjects


adequacy and clarity of information disclosure. Investigators are held respon-

sible for ascertaining that research participants have comprehended the infor-

mation they have been given about the proposed research. The purpose of

consent provisions is not protection from risk, as earlier federal policies

seemed to imply, but protection of autonomy and personal dignity, including

the personal dignity of incompetent persons incapable of acting autonomously.

The report went on to suggest that third parties should be encouraged to follow

the research as it proceeds and should be aware of the right to withdraw an

incompetent person from his or her research participation throughout the

process.

The principle of beneficence is an abstract norm that includes rules such as

“Do no harm,” “Balance benefits against risks,” and “Maximize possible benefits

and minimize possible harms.” This principle is satisfied in the research context

by refraining from intentionally causing injury and by ensuring that risks are

reasonable in relation to probable benefits. The National Commission required

that there be an array of data pertaining to benefits and risks and of alternative

ways of obtaining the benefits (if any) sought from involvement in research. It

demanded that, if possible, systematic and nonarbitrary presentations of risks

and benefits be made to research participants as part of the informed consent

process and that the assessment of risks and safeguards be considered by an

institutional review board (IRB) in weighing the justifiability of research pro-

tocols. The National Commission stated that participants ought not to be asked

or allowed to consent to more risk than is warranted by anticipated benefits and

that forms of risk incommensurate with participants’ previous experience

should not be imposed in the case of groups such as children, who might be

overburdened and possibly disturbed or terrified. However, the report recog-

nized that risks must be permitted during the course of many forms of research

in order for investigators to be positioned to distinguish harmful from beneficial

outcomes.

The principle of justice requires fairness in the distribution of both the

burdens and the benefits of research. The National Commission insisted that

this principle requires special levels of protection for vulnerable and disadvan-

taged parties. This principle demands that researchers first seek out and select

persons best prepared to bear the burdens of research (e.g., healthy adults) and

that they not offer research only to groups who have been repeatedly

targeted (e. g., mentally retarded children). The National Commission noted

that, historically, the burdens of research were placed heavily on the

economically disadvantaged, the very sick, and the vulnerable, owing to

their ready availability. This conclusion was based on published data, reports

in the media, public testimony to the National Commission, and some

onsite visits to places such as prisons. The overutilization of readily available,


and often compromised, segments of the U.S. population was a matter of deep

moral concern to the National Commission. The theme of justice and proper

selection of research participants was the Belmont Report’s way of saying that

because medical research is a social enterprise for the public good, it must be

accomplished in a broadly inclusive and participatory way. If participation in

research is unwelcome and falls on a narrow spectrum of citizens because of

their ready availability, then it is unwarranted. Likewise, the National

Commission recommended that persons who are already burdened by some

form of disability or institutionalization should not be asked to accept the

burdens of research—unless, as occurs in some cases, other participants

cannot be located or are otherwise inappropriate.

The Belmont Report includes not only these three abstract principles and their

analysis but also a moral view that moves modestly in the direction of an applied

research ethics. Just as there is a distinction between theoretical or general ethics

and applied or practical ethics, so the National Commission thought of the

Belmont Report as its statement of a brief theoretical framework. Because of its

practical and policy objectives, the explanation of the principles had a notably

applied character. Even so, the National Commission had no ambition to make

the report itself specific and practical for institutions that conduct research. This

objective was to be accomplished by the other 16 volumes on problems of

research ethics and recommendations for public policy that the National

Commission issued. The Belmont Report itself was intended to provide only a

general framework of basic principles.

National Commission members and staff were keenly aware that this frame-

work was too indeterminate by itself to decide practice or policy or to resolve

moral conflicts. The process of molding the general principles in the Belmont

Report so that they become sufficiently concrete is a process of reducing the

indeterminateness and abstractness of the principles to give them increased

action-guiding capacity. The report looks to educational institutions, professional

associations, government agencies, and IRBs to provide the more specific rules

and judgments required for research ethics.

The works of philosophers such as W. D. Ross and William Frankena were

often consulted in the drafting of the Belmont Report, but the moral principles

featured in the report should not be read as deriving from the writings of

philosophers. The Belmont Report made reference to values “generally accepted

in our cultural tradition” as the basis of its principles. These principles derived

from National Commission members’ understanding of social morality, and

these commissioners may have been confusing social morality with universal

morality. We will never know what the commissioners meant by our “tradi-

tion,” but the import of the Belmont principles is not to be tied to the unique

views of a particular tradition or nation. The National Commission apparently


conceived of its principles as universally valid norms. That is, the principles were

taken to be applicable to all contexts of human research, not merely to some

local region, such as an institution or a nation. The presumption is that no

responsible research investigator could conduct research without reference to

these principles; these principles form the core of any policy worthy of the name

“research ethics.”

WEAKNESSES, DEFICIENCIES, AND UNCLARITIES

Despite the Belmont Report’s wide acceptance, several issues have been or can be

raised about the adequacy of the Belmont principles. Here are six possible

problems.

1. The way the principles are delineated is arguably confused, especially

the principle of respect for persons. This principle appears to confusingly

blend two principles: a principle of respect for autonomy and a principle of

protecting and avoiding harm to incompetent (nonautonomous) persons. The

National Commission said that it was attempting to protect both autonomous

persons and those with “diminished autonomy,” that is, those who are incap-

able of self-determination. Both are persons, it said, and both are entitled to

protection.

The question is whether protections for persons who are incapable of self-

determination can be justified in any way other than by the principle of benefi-

cence. If not, the criticism goes, then the National Commission has adopted an

incoherent position in thinking that respect for persons and beneficence are

independent principles. Robert Veatch seems to both criticize and defend the

National Commission for this apparent confusion:

The Belmont Report offers a three-principle theory that uses the Kantian term:

“respect for persons.” It subsumes autonomy under this broader notion, but in

a strangeway it also subsumes the welfare of the incompetent under respect for

persons. This is hard to defend. Autonomy may be an element of a more

fundamental notion of respect for persons, but it seems that the duty to serve

the welfare of the incompetent is straightforwardly a part of the duty of

beneficence. Nevertheless, if respect for persons includes autonomy, it could

include other elements as well . . . [I myself include] principles other than

autonomy as part of respect for persons. If respect for persons includes only

respect for autonomy, then nonautonomous persons are left stranded . . .

Respecting persons must involve . . . Veracity . . . Fidelity to promises . . .

[and] Avoidance of killing.5


Veatch is suggesting that although the National Commission erred in its

explication of respect for persons, its general viewpoint could be reconstructed

and rendered viable.

However, there remains the problem of whether there is any meaningful way to

justify protections for persons who are incapable of self-determination except by

appeal to a principle such as beneficence or justice. What is it to respect an

incompetent person if it is not to act beneficently or justly in regard to the

person? Accordingly, the appropriate principle to determine one’s commitments

would seem to be either beneficence or justice, or both. It is the welfare of the

incompetent person that the National Commission seeks to protect, not the

personhood of the incompetent person. If this view is correct, then the National

Commission has adopted an incoherent position by thinking that respect for

persons and beneficence, and perhaps justice, are independent principles.

However, there is apparently a hidden target implicitly at work in the National

Commission’s choice of the language of “persons.” The principle of respect for

persons seems to be functioning to give incompetent persons amoral status that is

functionally equivalent to the moral status autonomous persons possess. That is,

the principle is functioning for the National Commission as a way of saying that

the absence of autonomy must not lead to the neglect, disadvantage, or lowered

esteem of incompetent individuals and those in vulnerable populations, and that

such persons have a right to appropriate third-party consent. In short, the

National Commission was attempting to handle issues of the moral status of

incompetent individuals through its interpretation of the principle of respect for

persons.

2. The National Commission was very concerned that using utilitarian justifi-

cations of research had become too easy in the biomedical world. The Nazi

experiments, Tuskegee, and the Jewish Chronic Disease Hospital cases all

seemed to have been driven by a very utilitarian view of (social) beneficence

that justified using humans on grounds of benefit to the broader public. However,

the National Commission itself has been accused of having inadequate internal

controls in its moral framework to protect research participants against abuse

when there is the promise of major benefit for society. Two National Commission

members, Robert Cooke and Robert Turtle, sternly criticized the National

Commission’s report on children on grounds that it had presented an unjustifi-

able utilitarian justification of research that placed children at undue risk.6,7

Whatever the merits of this criticism, the Belmont Report was written with the

intent to rectify what was perceived as an imbalance favoring social utility in the

justification of research. That is, a major purpose of the report was to properly

balance the interests of research participants with those of science and society.

Considerations of autonomy, justice, and risk control were set out to limit

utilitarian overbalancing and investigator discretion. However, it is doubtful


that the question of how best to control utilitarian balancing was ever resolved by

the National Commission.

3. Paradoxically, the Belmont Report, and the National Commission more

generally, can be criticized for being overly protective of research participants—

and, consequently, as insufficiently utilitarian to meet the needs of certain classes

of persons. The National Commission’s emphasis was on the protection of

persons from research injury. Research participation was conceived as a burden

that individuals accepted in order to advance the public good and that should be

distributed equitably. It is unclear why this assumption was so deep in the

National Commission’s work, but it can likely be explained by the atmosphere

of scandal that had emerged at the time. The nature and acceptability of the

public’s interest in research and the goals of research were little-explored matters

in the Belmont Report. The notion that research is not always viewed by partici-

pants as burdensome was underexamined.

The AIDS epidemic altered this mindset, perhaps forever. Whereas the Belmont

Report sought to protect research participants, AIDS activists sought not protec-

tion from, but inclusion in, the research process. They wanted justice and respect

for their autonomy, but not along the lines of justice and autonomy staked out in

the Belmont Report. They wanted to be able to choose unapproved drugs; con-

siderations of justice, they thought, should be used to allow them access to clinical

trials. To them, the Belmont principles could be interpreted as protectionist to the

point of excluding those who might benefit from research and from access to

potentially beneficial drugs. In the end, this push for inclusion in research and

broader access to the potential benefits of research altered the course of research

ethics.Whether this development constitutes an expansion in the scope and use of

the Belmont principles or a confrontation with these principles is debatable, but it

certainly reconfigured research ethics.4,8

4. The Belmont Report has also been criticized for its abstractness and inability

to resolve or otherwise treat practical moral problems. The report anticipated this

criticism and cautioned that its principles “cannot always be applied so as to

resolve beyond dispute particular ethical problems. The objective is [only] to

provide an analytical framework that will guide the resolution of ethical problems

arising from research involving human subjects.” The National Commission thus

warned readers that they should not expect to use Belmont principles as a checklist

of federal regulations or as guidelines like recipes in cooking. Nonetheless, several

critics have asked whether these principles are in any meaningful respect practical,

or even useful. The concern is that norms as general as the Belmont principles

underdetermine almost all moral judgments because they are abstract and contain

too little content to determine concrete judgments.

Danner Clouser and Bernard Gert have objected to the Belmont principles

and all related analytical frameworks, contending that principles function more


like chapter headings in a book than as directive rules and theories. Therefore,

receiving no directive guidance from the principle, anyone who is working on a

problem in bioethics is left free to deal with that principle in his or her own way

and may give it whatever meaning and significance he or she wishes. Consider

justice. The Belmont principle of justice in the selection of research participants

instructs a moral agent to be alert to various matters of justice, but does such a

general principle actually guide conduct? Other moral considerations besides the

principle(s) of justice, such as moral intuitions and ethical theories, may be

needed to do the real work of ethical reflection.9–11

This criticismmerits careful attention, but can any system of principles, rules,

or general guidelines escape this problem? Clouser and Gert maintain that

some general moral rules can provide deep and directive substance, but these

authors have had difficulty in clarifying and justifying this claim. Their point is

no doubt correct for unspecified principles, but all abstract principles and

rules will need some sort of specification in context to become directive. The

National Commission anticipated this problem. It did not advance the Belmont

principles as sufficient for resolving problems, but only as a starting point. The

National Commission noted that “other principles may be relevant” and that its

principles should be able to “serve as a basic justification for” more “particular

prescriptions and evaluations.”

5. The Belmont Report has also been faulted for another, but closely related,

reason: It gave no indication of how to prioritize or weigh its principles. Several

commentators assert that the National Commission should have argued that

one or more of its principles has priority—for example, that considerations of

respect for persons and justice take priority over considerations of social ben-

efit.5,7,12 These critics support a model of basic moral principles and protections

that cannot be violated under any circumstances, even if there is a clear and

substantial benefit for society. Such restrictions—often said to be deontological

in contrast to utilitarian—are analogous to constitutional rights that constrain

conduct and prohibit balancing of interests in various political and social

matters. Some who propose such an ordinal or priority ranking of principles

argue that it is the morally correct view, but others are more concerned to show

that the National Commission has not found a way out of situations in which its

own principles come into conflict. Some critics also point out that a priority

ranking of the sort they propose to mitigate conflict between beneficence and

either autonomy or social justice would allow the National Commission to

escape the accusation that its schema too readily invites utilitarian justifications

of research protocols.

Other commentators have argued that the National Commission was correct in

its assumption that the principles are more or less weighty depending on the

particular circumstances in which they are to be applied. Accounts of this sort are


often referred to as balancing theories. Such theories do not allow any principle in

the basic analytical framework to have an ordered priority over any other prin-

ciple. That is, there is neither a hierarchical ranking of principles nor an a priori

weighting of principles. When principles conflict, the balance of right over wrong

must be determined by assessing the weight of competing considerations as they

emerge in the circumstance. What agents ought to do is determined by what they

ought to do, all things considered. This appears to be the view presumed in the

Belmont Report, and it seems also to be the view currently accepted in federal

regulations for protocol review, waivers of consent, and the like.13–15

6. One former National Commission member, Albert Jonsen, and one

staff member, Stephen Toulmin, have jointly questioned whether the

National Commission actually used its framework of Belmont principles to

support or defend its own bioethical conclusions. They have argued that the

National Commission members wrote their documents as principlists, but

actually worked as casuists. This thesis is not a criticism of the National

Commission’s substantive work, only a methodological comment on the use

and limits of its principles. These authors hold that the National Commission’s

actual moral deliberations proceeded by the consideration of influential cases

rather than by appeal to universal principles, and they think, more generally,

that this paradigm of reasoning is the best method in bioethics. Jonsen and

Toulmin present this understanding of the National Commission’s work as

follows:

The one thing [individual Commissioners] could not agree on was why they

agreed. . . . Instead of securely established universal principles, . . . giving them

intellectual grounding for particular judgments about specific kinds of cases, it

was the other way around. . . . The locus of certitude in the Commissioners’

discussions . . . lay in a shared perception of what was specifically at stake in

particular kinds of human situations. . . . That could never have been derived

from the supposed theoretical certainty of the principles to which individual

Commissioners appealed in their personal accounts.

The point is that National Commission members reasoned by appeal to

particular cases and families of cases, and reached consensus through agreement

on central cases and on generalization from sets of cases. Principles were therefore

of lesser importance than readers might suppose when they read the Belmont

Report. Although the Belmont principles are important guiding ideals, they are, in

this estimate, overrated if revered for their practicality. Although Jonsen has

supported the moral principles delineated in the Belmont Report, he has also

maintained that in practical ethics, these principles must be interpreted and

specified by the force of examples and counterexamples that emerge from


experience with cases.18 From this perspective, the National Commission used

principles primarily as general guiding ideals.

ENDURING LEGACY AND INFLUENCE

The Belmont Report is one of the few documents to influence almost every sphere

of activity in bioethics: moral theory and general standards of research ethics,

government regulatory activity, bioethics consultation, and evenmedical practice.

Its influence has arguably been as extensive in practice as in theory.

Many interested in the role of moral theory and principles in bioethics have

honored the Belmont Report for its framework of principles, even if those

principles have not been widely analyzed in this literature. As Dan Brock has

observed, “The Belmont Report . . . had great impact on bioethics because it

addressed the moral principles that underlay the various reports on particular

aspects of research.”19 Brock is noting the influence of the idea that a body of

principles can be used to frame and discuss a wide range of practical moral

problems.

In federal regulatory oversight and law, the Belmont Report has at times

assumed a near canonical role. The Advisory Committee on Human Radiation

Experiments noted in 1995 that

Many conditions coalesced [historically] into the framework for the

regulation of the use of human subjects in federally funded research that is

the basis for today’s system. . . . [T]his framework is undergirded by the three

Belmont principles. The federal regulations and the conceptual framework

built on the Belmont principles became so widely adopted and cited that it

might be argued that their establishment marked the end of serious

shortcomings in federal research ethics policies.20

Similarly, an Institute ofMedicine report, issued by its Committee on Assessing

the System for Protecting Human Research Participants, stated in 2002 that “The

ethical foundations of research protections in the United States can be found in

the three tenets identified in the Belmont Report.”21 Moreover, the Belmont

principles found their way into every document the National Commission pub-

lished, and these became the backbone of federal law. From this perspective, as

Christine Grady has observed, “probably the single most influential body in the

United States involved with the protection of human research subjects was the

National Commission.”22

The legacy of Belmont may be most enduring in areas of practice. Federal

regulations require that all institutions receiving federal funds for research


espouse a statement of principles for the protection of human research partici-

pants. Virtually all such institutions have subscribed to the Belmont principles as

the basis of their efforts to assess research protocols from an ethical point of view.

Professional associations, too, have widely recognized the authority and historical

significance of the Belmont principles.23 Eric Cassell has even argued that the

Belmont principles have “permeated clinical medicine” as extensively as they have

medical research.24His claim is that the Belmont principles were a significant force

in a broad cultural shift in medicine toward a reworking of the relationship

between doctor and patient.

Whatever the influence and enduring legacy of Belmont, it is not clear

that scientists who are involved today in research with humans are any more

familiar with the Belmont principles than their predecessors of several decades

ago were familiar with documents such as the Nuremberg Code. When the

National Commission deliberated, it seemed to some observers that the general

system of protecting human research participants in the United States was in

need of serious repair, that research investigators were not educated about

research ethics, and that participants were not adequately protected. To some

observers, the system seems today caught in a similar state of disrepair. From

1997 to 2002, a large number of hearings, bills, and reports by official, prestigious

government bodies and government-mandated bodies in the United States

concluded that the system of IRB review and the practice of informed con-

sent—the core of research ethics established by the National Commission—are

seriously defective.8,21

The National Commission and its Belmont Reportmay have succeeded both in

“resolving” somemajor problems of research ethics and in bringing “oversight” to

the research context, as historian David Rothman has claimed,25 but this fix may

have been temporary and time-bound. Today, the Belmont principles may be

more revered than they are understood and practiced.

Notes

1. The National Commission for the Protection of Human Subjects of Biomedical and

Behavioral Research. The Belmont Report: Ethical Principles and Guidelines for the Protection

of Human Subjects of Research. Washington, DC: Department of Health, Education and

Welfare; DHEW Publication OS 78-0012 1978. Available at: http://www.hhs.gov/ohrp/

humansubjects/guidance/belmont.htm.

2. Department of Health and Human Services, National Institutes of Health, and Office

for Human Research Protections. The Common Rule, Title 45 (Public Welfare), Code of

Federal Regulations, Part 46 (Protection of Human Subjects). Available at: http://www.hhs.

gov/ohrp/humansubjects/guidance/45cfr46.htm.

3. Jones JH. Bad Blood, 2nd ed. New York: The Free Press; 1993.


http://www.hhs.gov/ohrp/humansubjects/guidance/belmont.htm

http://www.hhs.gov/ohrp/humansubjects/guidance/belmont.htm

http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm

http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm

4. Levine C. “Changing views of justice after Belmont: AIDS and the inclusion

of ‘vulnerable’ subjects.” In: Vanderpool HY, ed. The Ethics of Research Involving

Human Subjects: Facing the 21st Century. Frederick, MD: University Publishing Group;

1996: 105–126.

5. Veatch RM. “Resolving conflicts among principles: Ranking, balancing, and speci-

fying.” Kennedy Institute of Ethics Journal 1995; 5: 199–218.

6. National Commission for the Protection of Human Subjects of Biomedical and

Behavioral Research. Archived Materials 1974–78. Transcript of the Meeting Proceedings

(for discussion of the Belmont Paper at the following meetings: February 11–13, 1977;

July 8–9, 1977; April 14–15, 1978; and June 9–10, 1978), Kennedy Institute Library,

Georgetown University, Washington, DC.

7. Marshall E. “Does the moral philosophy of the Belmont Report rest on a mistake?”

IRB: A Review of Human Subjects Research 1986; 8(6): 5–6.

8. Moreno JD. “Goodbye to all that: The end of moderate protectionism in human

subjects research.” Hastings Center Report 2001; 31(3): 9–17.

9. Clouser KD, Gert B. “Morality vs. principlism.” In: Gillon R, Lloyd A, eds. Principles

of Health Care Ethics. London, England: John Wiley and Sons; 1994: 251–266.

10. Gert B, Culver CM, Clouser KD. Bioethics: A Return to Fundamentals. New

York: Oxford University Press; 1997: 72–75.

11. Clouser KD, Gert B. “A critique of principlism.” Journal of Medicine and

Philosophy 1990; 15: 219–236.

12. Veatch RM. “From Nuremberg through the 1990s: The priority of autonomy.” In:

Van-derpool HY, ed. The Ethics of Research Involving Human Subjects: Facing the 21st

Century. Frederick, MD: University Publishing Group; 1996: 45–58.

13. Ackerman TF. “Choosing between Nuremberg and the National Commission:

The balancing of moral principles in clinical research.” In Vanderpool HY, ed. The

Ethics of Research Involving Haman Subjects: Facing the 21st Century. Frederick, MD:

University Publishing Group; 1996, 83–104.

14. Beauchamp TL, Childress JF. Principles of Biomedical Ethics, 6th ed. New York:

Oxford University Press; 2009: Chapters 1, 10.

15. Jonsen AR. “The weight and weighing of ethical principles.” In: Vanderpool HY.

The Ethics of Research Involving Human Subjects: Facing the 21st Century. Frederick, MD:

University Publishing Group; 1996: 64–65.

16. Jonsen AR, Toulmin S. The Abuse of Casuistry. Berkeley, CA: University of California

Press; 1988: 16–19.

17. Toulmin S. “The National Commission on human experimentation: Procedures

and outcomes.” In: Engelhardt HT Jr., Caplan A, eds. Scientific Controversies: Case Studies

in the Resolution and Closure of Disputes in Science and Technology. New York: Cambridge

University Press; 1987: 599–613.

18. Jonsen AR. “Casuistry.” In: Sugarman J, Sulmasy DP, eds. Methods of

Bioethics. Washington, DC: Georgetown University Press; 2001: 112–113.

19. Brock D. “Public policy and bioethics.” In: Reich WT, ed. Encyclopedia of Bioethics,

2nd ed. New York: Macmillan Reference; 1995: 2181–2188.


20. Advisory Committee on Human Radiation Experiments. Final Report of the Advisory

Committee on Human Radiation Experiments. New York: Oxford University Press; 1996.

21. Institute of Medicine, Committee on Assessing the System for Protecting

Human Research Participants (Federman DE, Hanna KE, Rodriguez LL, eds.).

Responsible Research: A Systems Approach to Protecting Research Participants.

Washington, DC: National Academies Press; 2002.

22. Grady C. The Search for an AIDS Vaccine: Ethical Issues in the Development and

Testing of a Preventive HIV Vaccine. Bloomington, IN: Indiana University Press; 1995: 42.

23. Pincus HA, Lieberman JA, Ferris S. Ethics in Psychiatric Research. Washington,

DC: American Psychiatric Publishing, Inc.; 1999.

24. Cassell EJ. “The principles of the Belmont Report revisited: How have respect for

persons, beneficence, and justice been applied to clinical medicine?” Hastings Center

Report 2000; 30(4): 12–21.

25. Rothman DJ. “Research, human: Historical aspects.” In: Reich WT, ed. Encyclopedia

of Bioethics, 2nd ed. New York: Macmillan Reference; 1995: 2256.


Part II

PRINCIPLISM AND PRACTICE

3

THE FOUR PRINCIPLES APPROACH TO

HEALTH CARE ETHICS

My objective in this essay is to explain the so-called “four principles” approach

and to explain the philosophical and practical roles these principles play. I start

with a brief history and then turn to the four-principles framework, its

practical uses, and philosophical problems of making the abstract framework

specific.

THE ORIGINS OF PRINCIPLES IN HEALTH CARE ETHICS

Prior to the early 1970s, there was no firm ground in which a commitment to

principles or even ethical theory could take root in biomedical ethics. This is not

to say that physicians and researchers had no principled commitments to

patients and research subjects. They did, but moral principles, practices, and

virtues were rarely discussed. The health care ethics outlook in Europe and

America was largely that of maximizing medical benefits and minimizing risks

of harm and disease. The Hippocratic tradition had neglected many problems of

truthfulness, privacy, justice, communal responsibility, the vulnerability of

research subjects, and the like.1 Views about ethics had been confined largely

to the perspectives of those in the professions of medicine, public health, and

nursing. No sustained work combined concerns in ethical theory and the health

care fields.

Principles that could be understood with relative ease by the members of

various disciplines figured prominently in the development of biomedical ethics

during the 1970s and early 1980s. Principles were used to present frameworks of

evaluative assumptions so that they could be used by, and readily understood by,

people with many different forms of professional training. The distilled morality

found in principles gave people a shared and serviceable group of general norms for

analyzing many types ofmoral problems. In some respects, it could even be claimed

that principles gave the embryonic field of bioethics a shared method for attacking

its problems, and this gave some minimal coherence and uniformity to bioethics.

There were two primary sources of the early interest in principles in biomedical

ethics. The first was the Belmont Report (and related documents) of the National

Commission for the Protection of Human Subjects,2 and the second was the book

entitled Principles of Biomedical Ethics, which I coauthored with James

F. Childress. I here confine discussion to the latter.

Childress and I began our search for the principles of biomedical ethics in

1975. In early 1976, we drafted the main ideas for the book, although only later

would the title Principles of Biomedical Ethics be placed on it.3 Our goal was to

develop a set of principles suitable for biomedical ethics. One of our proposals

was that health care’s traditional preoccupation with a beneficence-based model

of health care ethics be augmented by an autonomy model, while also incorpor-

ating a wider set of social concerns, particularly those focused on social justice.

The principles are understood as standards of conduct on which many other

moral claims and judgments depend. A principle, then, is an essential norm in a

system of moral thought and one that is basic for moral reasoning. More specific

rules for health care ethics can be formulated by reference to these four princi-

ples, but neither rules nor practical judgments can be straightforwardly deduced

from the principles.

THE FRAMEWORK OF PRINCIPLES

The principles in our framework have always been grouped under four general

categories: (1) respect for autonomy (a principle requiring respect for the deci-

sion-making capacities of autonomous persons), (2) nonmaleficence (a principle

requiring not causing harm to others), (3) beneficence (a group of principles

requiring that we prevent harm, provide benefits, and balance benefits against

risks and costs), and (4) justice (a group of principles requiring appropriate

distribution of benefits, risks, and costs fairly). I will concentrate on an explication

of each of the principles and how they are to be understood collectively as a

framework.

36 P R I N C I P L I S M A N D P R A C T I C E

Respect for Autonomy

Respect for autonomy is rooted in liberal moral and political traditions of the

importance of individual freedom and choice. In moral philosophy, personal

autonomy refers to personal self-governance: personal rule of the self by adequate

understanding while remaining free from controlling interferences by others and

from personal limitations that prevent choice. Autonomy means freedom from

external constraint and the presence of critical mental capacities such as under-

standing, intending, and voluntary decision making.4 The autonomous indivi-

dual acts freely in accordance with a self-chosen plan, analogous to the way an

independent government manages its territories and sets its policies. A person

of diminished autonomy, by contrast, is in some respect controlled by others or

incapable of deliberating or acting on the basis of his or her desires and plans.

To respect an autonomous agent is to recognize with due appreciation that

person’s capacities and perspectives, including his or her right to hold certain

views, to make certain choices, and to take certain actions based on personal

values and beliefs. The moral demand that we respect the autonomy of persons

can be expressed as a principle of respect for autonomy that states both a negative

obligation and a positive obligation. As a negative obligation, autonomous

actions should not be subjected to controlling constraints by others. As a positive

obligation, this principle requires respectful treatment in informational

exchanges and in other actions that foster autonomous decision making.

Many autonomous actions could not occur without others’ material coopera-

tion in making options available. Respect for autonomy obligates professionals in

health care and research involving human subjects to disclose information, to

probe for and ensure understanding and voluntariness, and to foster adequate

decisionmaking. True respect requires more thanmere noninterference in others’

personal affairs. It includes, at least in some contexts, building up or maintaining

others’ capacities for autonomous choice while helping to allay fears and other

conditions that destroy or disrupt their autonomous actions. Respect, on this

account, involves acknowledging the value and decision-making rights of persons

and enabling them to act autonomously, whereas disrespect for autonomy

involves attitudes and actions that ignore, insult, demean, or are inattentive to

others’ rights of autonomy.

Many issues in professional ethics concern failures to respect a person’s

autonomy, ranging from manipulative underdisclosure of pertinent information

to nonrecognition of a refusal of medical interventions. For example, in the debate

over whether autonomous, informed patients have the right to refuse medical

interventions, the principle of respect for autonomy suggests that an autonomous

decision to refuse interventions must be respected. Although it was not until the

late 1970s that serious attention was given to the rights of patients to refuse, this is

T H E F O U R P R I N C I P L E S A P P R O A C H T O H E A L T H C A R E E T H I C S 37

no reason for thinking that respect for autonomy as now understood is a newly

added principle in our moral perspective. It simply means that the implications of

this principle were not widely appreciated until recently.5

Controversial problems with the principle of respect for autonomy, as with all

moral principles, arise when we must interpret its significance for particular

contexts, determine precise limits on its application, and decide how to handle

situations in which it conflicts with other moral principles. Many controversies

involve questions about the conditions under which a person’s right to autono-

mous expression demands actions by others, as well as questions about the

restrictions society may rightfully place on choices by patients or subjects when

these choices conflict with other values. If restriction of the patient’s autonomy is

in order, the justification will always rest on some competingmoral principle such

as beneficence or justice.

Nonmaleficence

Physicians have long avowed that they are obligated to avoid doing harm to their

patients. Among the most quoted principles in the history of codes of health care

ethics is themaxim primum non nocere: “Above all, do no harm.” British physician

Thomas Percival furnished the first developed modern account of health care

ethics. He maintained that a principle of nonmaleficence fixes the physician’s

primary obligations and triumphs even over respect for the patient’s autonomy in

a circumstance of potential harm to patients:

To a patient . . . who makes inquiries which, if faithfully answered, might

prove fatal to him, it would be a gross and unfeeling wrong to reveal the

truth. His right to it is suspended, and even annihilated; because . . . it would

be deeply injurious to himself, to his family, and to the public. And he has the

strongest claim, from the trust reposed in his physician, as well as from the

common principles of humanity, to be guarded against whatever would be

detrimental to him.6

Some basic rules in the common morality are requirements to avoid causing a

harm. They include rules such as do not kill; do not cause pain; do not disable; do

not deprive of pleasure; do not cheat; and do not break promises.7 Similar but

more specific prohibitions are found across the literature of biomedical ethics,

each grounded in the principle that intentionally or negligently caused harm is a

fundamental moral wrong.

Numerous problems of nonmaleficence are found in health care ethics today,

some involving blatant abuses of persons and others involving subtle and


unresolved questions. Blatant examples of failures to act nonmaleficently are

found in the use of physicians to classify political dissidents as mentally ill,

thereafter treating them with harmful drugs and incarcerating them with

insane and violent persons.8 More subtle examples are found in the use of

medications for the treatment of aggressive and destructive patients. These

common treatment modalities are helpful to many patients, but they can be

harmful to others.

A provocative question about nonmaleficence and physician ethics has

been raised by Paul S. Appelbaum in an investigation of “the problem of

doing harm” through testimony in criminal contexts and civil litigation—

for example, by omitting information in the context of a trial, after which

a more severe punishment is delivered to the person than likely would

have been delivered. Appelbaum presents the generic problem as one of

nonmaleficence:

If physicians are committed to doing good and avoiding harm, how can they

participate in legal proceedings from which harmmay result? If, on the other

hand, physicians in court abandon medicine’s traditional ethical principles,

how do they justify that deviation? And if the obligations to do good and

avoid harm no longer govern physicians in the legal setting, what alternative

principles come into play? . . . Are physicians in general bound by the

principles of beneficence and nonmaleficence?9

Beneficence

The physician who professes to “do no harm” is not usually interpreted as

pledging never to cause harm, but rather to strive to create a positive balance of

goods over inflicted harms. Those engaged in medical practice, research, and

public health know that risks of harm presented by interventions must often be

weighed against possible benefits for patients, subjects, and the public. Here we

see the importance of beneficence as a principle beyond the scope of

nonmaleficence.

In ordinary English, the term beneficence connotes acts of mercy, kindness,

charity, love, and humanity. In its most general meaning, it includes all forms of

action intended to benefit other persons. In health care ethics beneficence com-

monly refers to an action done to benefit others, whereas benevolence refers to the

character trait or virtue of being disposed to act for the benefit of others. The

principle of beneficence refers to amoral obligation to act for the benefit of others.

No demand is more important in taking care of patients: The welfare of patients is

medicine’s context and justification. Beneficence has long been treated as a


foundational value—and sometimes as the foundational value10—in health care

ethics.

The principle of beneficence requires us to help others further their important

and legitimate interests, often by preventing or removing possible harms. This

principle includes rules such as “maximize possible benefits and minimize

possible harms” and “balance benefits against risks.” Many duties in medicine,

nursing, public health, and research are expressed in terms of a positive obliga-

tion to come to the assistance of those in need of treatment or in danger of

injury. The harms to be prevented, removed, or minimized are the pain,

suffering, and disability of injury and disease. The range of benefits that might

be considered relevant is broad. It could even include helping patients find

appropriate forms of financial assistance and helping them gain access to health

care or research protocols. Sometimes the benefit is for the patient, at other

times for society.

Some writers in health care ethics suggest that certain duties not to injure others

are more compelling than duties to benefit them. They point out that we do not

consider it justifiable to kill a dying patient in order to use the patient’s organs to

save two others, even though benefits would be maximized, all things considered.

The obligation not to injure a patient by abandonment has been said to be

stronger than the obligation to prevent injury to a patient who has been aban-

doned by another (under the assumption that both are moral duties). Despite the

attractiveness of these notions that there is a hierarchical ordering rule, Childress

and I reject such hierarchies on grounds that obligations of beneficence do, under

many circumstances, outweigh those of nonmaleficence. A harm inflicted by not

avoiding causing it may be negligible or trivial, whereas the harm that beneficence

requires we prevent may be substantial. For example, saving a person’s life by a

blood transfusion clearly justifies the inflicted harm of venipuncture on the blood

donor. One of the motivations for separating nonmaleficence from beneficence is

that these principles themselves come into conflict. Since the weights of the two

principles can vary, there can be no mechanical decision rule asserting that one

obligation must always outweigh the other.

Perhaps the major theoretical problem about beneficence is whether the prin-

ciple generates general moral duties that are incumbent on everyone—not

because of a professional role, but because morality itself makes a general

demand of beneficence. Many analyses of beneficence in ethical theory (most

notably utilitarianism11) seem to demand severe sacrifice and extreme generosity

in the moral life—for example, giving a kidney for transplantation or donating

bone marrow to a stranger. However, many moral philosophers have argued that

such beneficent action is virtuous and a moral ideal, but not an obligation;

therefore, there is no principle of beneficence of the sort proclaimed in the

four-principles approach.


I agree that the line between what is required and what is not required by the

principle is difficult to draw, and that drawing a precise line independent of

context is impossible. I do not agree, however, with the radical view that there are

no obligations of beneficence—both general and specific obligations. I return to

this problem of weighing, judging, and specifying later in this essay in a discussion

of the notion of prima facie duties.

Justice

Every civilized society is a cooperative venture structured by moral, legal, and

cultural principles of justice that define the terms of cooperation. A person in any

such society has been treated justly if treated according to what is fair, due, or

owed. For example, if equal political rights are due to all citizens, then justice is

done when those rights are accorded. The more restricted notion of distributive

justice refers to fair, equitable, and appropriate distribution in society. Usually this

term refers to the distribution of primary social goods, such as economic goods

and fundamental political rights, but burdens are also within its scope. Paying for

forms of national health insurance is a distributed burden; medical-welfare checks

and grants to do research are distributed benefits.

There is no single principle of justice in the four-principles approach.

Somewhat like principles under the heading of beneficence, there are several

principles, each requiring specification in particular contexts. But common to

almost all theories of justice—and accepted in the four-principles approach—is

theminimal (formal) principle that like cases should be treated alike, or, to use the

language of equality, equals ought to be treated equally and unequals unequally.

This elementary principle, or formal principle of justice, states no particular

respects in which people ought to be treated. It merely asserts that whatever

respects are relevant, if persons are equal in those respects, they should be treated

alike. Thus, the formal principle of justice does not tell us how to determine

equality or proportion in these matters, and it lacks substance as a specific guide

to conduct.

Many controversies about justice arise over what should be considered the

relevant characteristics for equal treatment. Principles that specify these relevant

characteristics are often said to be “material” because they identify relevant

properties for distribution. Childress and I take account of the fact that philoso-

phers have developed diverse theories of justice that provide sometimes conflicting

material principles. We try to show that there are some merits in egalitarian

theories, libertarian theories, and utilitarian theories; and we defend a mixed use

of principles in these theories. We regard these three theories of justice as

appropriately capturing some of our traditional convictions about justice,


and we think that they can all be tapped as resources that will help to produce a

coherent conception of justice.

However, many issues of justice in health care ethics are not easily framed in the

context of traditional principles and abstract moral theories.12 For example, some

basic issues in health care ethics in the last three decades center on special levels of

protection and aid for vulnerable and disadvantaged parties in health care sys-

tems. These issues cut across clinical ethics, public health ethics, and research

ethics. The four-principles approach tries to deal with several of these issues,

without producing a grand theory for resolving all issues of justice. For example,

we address issues in research ethics about whether research is permissible with

groups who have been repeatedly used as research subjects, though the advantages

of research are calculated to benefit all in society. We argue that since medical

research is a social enterprise for the public good, it must be accomplished in a

broadly inclusive and participatory way, and we try to specify the commitments of

such generalizations. In this way, we incorporate principles of justice but do not

produce a general theory of justice.

THE CHOICE OF FOUR PRINCIPLES AND THE EVOLUTION

OF THE THEORY

The choice of our four types of moral principle as the framework for moral

decision making in bioethics derives in part from professional roles and tradi-

tions. As I noted earlier, the obligations and virtues of health professionals have

for centuries, as found in codes and learned writings on ethics, been framed by

professional commitments to provide medical care and to protect patients from

disease, injury, and system failure. Our principles build on this tradition, but they

also significantly depart from it by including parts of morality that traditionally

have been neglected in health care ethics, attending especially to the traditional

neglect of principles of respect for autonomy and justice. All four types of

principles are needed to provide a comprehensive framework for biomedical

ethics, but this general framework is abstract and spare until it has been further

specified—that is, interpreted and adapted for particular circumstances.

Principles of Biomedical Ethics has evolved appreciably since the first edition in

its understanding of abstractness and the demands of particular circumstances.

This is not because the principles have changed, but because over the years

Childress and I have altered some of our views about the grounding of

the principles and about their practical significance. Two major changes deserve

special attention. The first is our development of the idea that the four principles

are already embedded in public morality—a universal common morality—and

are presupposed in the formulation of public and institutional policies.


The second is our adoption of Henry Richardson’s account of the specification of

moral norms. These supplements to our theory and their significance will be

discussed in the next two sections.

The Centrality of the Common Morality

The source of the four principles is what Childress and I call the common morality

(a view only incorporated at the point of the third edition of Principles, following

the language of Alan Donagan). The commonmorality is applicable to all persons

in all places, and all human conduct is rightly judged by its standards. The

following are examples of standards of action (here rules of obligation) in the

common morality: (1) don’t kill, (2) don’t cause pain or suffering to others, (3)

prevent evil or harm from occurring, (4) rescue persons in danger, (5) tell the

truth, (6) nurture the young and dependent, (7) keep your promises, (8) don’t

steal, (9) don’t punish the innocent, and (10) treat all persons with equal moral

consideration.

Why have such norms become parts of a common morality, whereas other

norms have not? To answer this question, I start with an assumption about the

primary goal—that is, objective—of the social institution of morality. This

objective is to promote human flourishing by counteracting conditions that

cause the quality of people’s lives to worsen. The goal is to prevent or limit

problems of indifference, conflict, suffering, hostility, scarce resources, limited

information, and the like. Centuries of experience have demonstrated that the

human condition tends to deteriorate into misery, confusion, violence, and

distrust unless norms of the sort just listed—the norms of the common mor-

ality—are observed. When complied with, these norms lessen human misery and

preventable death. It is an overstatement to maintain that all of these norms are

necessary for the survival of a society (as some philosophers and social scientists

have maintained13), but it is not too much to claim that these norms are necessary

to ameliorate or counteract the tendency for the quality of people’s lives to worsen

or for social relationships to disintegrate.14

These norms are what they are, and not some other set of norms, because they

have proven over time that their observance is essential to realize the objectives of

morality. What justifies them is that they achieve the objectives of morality, not

the fact that they are universally shared across cultures. It is conceivable, of course,

that the set of norms that is shared universally is not the same set of norms as the

set pragmatically justified by their conformity to the objectives of morality. I agree

that if another set of norms would better serve the objectives of morality, then that

set of norms ought to displace the norms currently in place. However, there are no

clear candidates as alternatives to these norms.


What Childress and I call “principles” are the most general and basic norms of

the common morality. In Principles of Biomedical Ethics, we devote an entire

chapter to each principle in the attempt to explain its nature, content, specifica-

tion, and the like. The assumption behind the argument in each chapter is that our

framework of four principles should incorporate and articulate the most general

values of the common morality.

Our framework encompasses several types of moral norms, including not only

principles but also rules, rights, and moral ideals. We treat principles as the most

general and comprehensive norms, but we make only a loose distinction between

rules and principles. Rules, we argue, are more precise and practical guides to

action that depend on the more general principles for their justification. We

defend several types of rules, all of which should be viewed as specifications of

principles. These include substantive rules (e.g., truth telling, confidentiality, and

privacy rules), authority rules (e.g., rules of surrogate authority and rules of

professional authority), and procedural rules (e.g., rules for determining elig-

ibility for organ transplantation and rules for reporting grievances to higher

authorities).

The Prima Facie Character of Principles and Rules

These principles and rules (or other norms in the common morality) can in some

circumstances be justifiably overridden by other moral norms with which they

conflict. For example, we might justifiably not tell the truth in order to prevent

someone from killing another person, and we might justifiably disclose confiden-

tial information about a person in order to protect the rights of another person.

Principles, duties, and rights are not absolute (or unconditional) merely because

they are universal. There are exceptions to all principles, each of which is merely

presumptive in force.

Oxford philosopher W. D. Ross developed a theory that has been influential on

Principles since the first edition. Ross’s theory is intended to assist in resolving

problems of conflict between principles. His views are based on an account of

prima facie duties, which he contrasts with actual duties. A prima facie duty is one

that is always to be acted upon unless it conflicts on a particular occasion with

another duty. One’s actual duty, by contrast, is determined by an examination of

the respective weights of competing prima facie duties in particular situations.

When principles contingently conflict, no supreme principle is available—in the

four-principles approach—to determine an overriding obligation. Therefore,

discretionary judgment becomes an inescapable part of moral thinking in our

approach.


Here is an example. A physician has confidential information about a patient

who is also an employee in the hospital where the physician practices. The

employee is seeking advancement in a stress-filled position, but the physician

has good reason to believe this advancement would be devastating for both

the employee and the hospital. The physician has duties of confidentiality,

nonmaleficence, and beneficence in these circumstances: Should the physician

break confidence? Could thematter be handled bymaking thin disclosures only to

the hospital administrator and not to the personnel office? Can such disclosures

be made consistent with one’s general commitments to confidentiality?

Addressing these questions through a process of moral justification is required

to establish one’s actual duty in the face of these conflicts of prima facie duties.

Once we acknowledge that all general principles have exceptions, we are free to

view every moral conclusion that is supported by a principle and every principle

itself as subject to modification or reformulation. Change of this sort is to be

accomplished through specification, the means by which principles come to have

their practical value.

The Specification of Norms and Moral Coherence

To say that principles have their origins in and find support in the common

morality and in traditions of health care is not to say that their appearance in a

developed system of biomedical ethics is identical to their appearance in the

traditions from which they spring. Prima facie principles underdetermine moral

judgments because there is too little content in abstract principles to determine

concrete outcomes. Every norm and theory contains regions of indeterminacy

that need reduction through further development of their commitments in the

system, augmenting them with a more specific moral content. I turn, then, to

these questions: How does the prima facie conception of principles work in

practical bioethics? How are general principles to reach down to concrete policies?

How does one fill the gap between abstract principles and concrete judgments?

The Method of Specification

The answer is that principles must be specified to suit the needs and demands of

particular contexts, thus enabling principles to overcome their lack of content and

to handle moral conflict. Specification is a process of reducing the indeterminate-

ness of abstract norms and providing them with specific action-guiding content.15

For example, without further specification, “do no harm” is too abstract to help in


thinking through problems such as whether physicians may justifiably hasten the

death of patients. The general norm has to be specified for this particular context.

Specification is not a process of either producing or defending general norms

such as those in the common morality; it assumes that they are available.

Specifying the norms with which one starts (whether those in the common

morality or norms that were previously specified) is accomplished by narrowing

the scope of the norms, not by explaining what the general normsmean. The scope

is narrowed, as Henry Richardson puts it, by “spelling out where, when, why, how,

by what means, to whom, or by whom the action is to be done or avoided.”16 For

example, the norm that we are obligated to “respect the autonomy of persons”

cannot, unless specified, handle complicated problems of what to disclose or

demand in clinical medicine and research involving human subjects. A definition

of “respect for autonomy” (as, say, “allowing competent persons to exercise their

liberty rights”) might clarify one’s meaning in using the norm, but would not

narrow the general norm or render it more specific.

Specification adds content to general norms. For example, one possible speci-

fication of “respect the autonomy of persons” is “respect the autonomy of

competent patients when they become incompetent by following their advance

directives.” This specification will work well in some medical contexts, but will

not be adequate in others, thus necessitating additional specification. Progressive

specification can continue indefinitely, gradually reducing the conflicts that

abstract principles themselves cannot resolve. However, to qualify all along the

way as a specification, a transparent connection must always be maintained to the

initial norm that gives moral authority to the resulting string of norms.

Now we come to a critical matter about particular moralities, by contrast to the

common morality. There is always the possibility of developing more than one

line of specification of a norm when confronting practical problems and moral

disagreements. It is simply part of the moral life that different persons and groups

will offer different (sometimes conflicting) specifications, potentially creating

multiple particular moralities. On any problematic issue (such as abortion,

animal research, aid in disaster relief, health inequities, and euthanasia), com-

peting specifications are likely to be offered by reasonable and fair-minded

parties, all of whom are committed to the common morality. We cannot hold

persons to a higher standard than to make judgments conscientiously in light of

the relevant basic and specified norms while attending to the available factual

evidence. Conscientious and reasonable moral agents will understandably dis-

agree with equally conscientious persons over moral weights and priorities in

circumstances of a contingent conflict of norms.

Nothing in the model of specification suggests that we can always eliminate

circumstances of intractable conflicting judgments. However, we should always

try to do so by justifying the specification we put forward. This suggests that


specification as a method needs to be connected to a model of justification that

will support some specifications and not others. Only brief attention can be paid

here to this difficult philosophical problem.

Justifying Specifications Using the Method of Coherence

A specification is justified, in the four-principles approach, if and only if it

maximizes the coherence of the overall set of relevant, justified beliefs. These

beliefs could include empirically justified beliefs, justified basic moral beliefs, and

previously justified specifications. This is a version of so-called wide reflective

equilibrium.17Nomatter how wide the pool of beliefs, there is no reason to expect

that the process of rendering norms coherent by specification will come to an end

or be perfected. Particular moralities are, from this perspective, continuous works

in progress—a process rather than a finished product. There is no reason to think

that morality can be rendered coherent in only one way through the process of

specification. Many particular moralities present coherent ways to specify the

common morality. Normatively, we can demand no more than that agents

faithfully specify the norms of the common morality with an attentive eye to

overall coherence.

The following are some of the criteria for a coherent (and therefore, according

to this model, justified) set of ethical beliefs: consistency (the avoidance of

contradiction), argumentative support (explicit support for a position with

reasons), intuitive plausibility (the feature of a norm or judgment being secure

in its own right), compatibility or coherence with reasonable nonmoral beliefs

(in particular, coherence with available empirical evidence), comprehensiveness

(the feature of covering the entire moral domain, or as much of it as possible), and

simplicity (reducing the number of moral considerations to the minimum pos-

sible without sacrifice in terms of the other criteria).18

CONCLUSION

I have explained, and argued in defense of, what has often been called the four-

principles approach to biomedical ethics and is now increasingly called princip-

lism.19 The four clusters of principles derive from both considered judgments in

the common morality and enduring and valuable parts of traditions of health

care. Health care ethics has often been said to be an “applied ethics,” but this

metaphor may be more misleading than helpful. It is rarely the case that we can

directly apply a principle to resolve a tough problem. We will almost always be

engaged in collecting evidence, reasoning, and specifying general principles. This


is how problems should be treated and how progress can be made in health care

ethics. From this perspective, the four principles form only a starting point—the

point where the practical work begins.

Notes

1. Albert R. Jonsen, The Birth of Bioethics (New York: Oxford University Press, 1998), 3ff;

Edmund D. Pellegrino and David C. Thomasma, The Virtues in Medical Practice (New

York: Oxford University Press, 1993), 184–189.


Behavioral Research, The Belmont Report: Ethical Principles and Guidelines for the

Protection of Human Subjects of Research (Washington, DC: DHEW Publication OS

78-0012 1978); James F. Childress, Eric M. Meslin, Harold T. Shapiro, eds. Belmont

Revisited: Ethical Principles for Research with Human Subjects (Washington, DC:

Georgetown University Press, 2005).

3. Tom L. Beauchamp and James F. Childress, Principles of Biomedical Ethics, 1st edition

(New York: Oxford University Press, 1979).

4. H. Tristram Engelhardt, Jr., The Foundations of Bioethics (New York: Oxford

University Press, 1996), 2nd edition; Jay Katz, The Silent World of Doctor and Patient

(New York: The Free Press, 1984); James F. Childress, “The Place of Autonomy in

Bioethics,” Hastings Center Report 20 (January/February 1990): 12–16; Rebecca Kukla,

“Conscientious Autonomy: Displacing Decisions in Health Care,” Hastings Center Report

(March-April 2005): 34–44.

5. Ruth R. Faden and Tom L. Beauchamp, A History and Theory of Informed Consent

(New York: Oxford, 1986).

6. Thomas Percival, Medical Ethics; or a Code of Institutes and Precepts, Adapted to the

Professional Conduct of Physicians and Surgeons (Manchester: S. Russell, 1803), 165–166.

Percival’s work was the pattern for the AmericanMedical Association’s (AMA) first code of

ethics in 1847.

7. Compare Bernard Gert, Morality: Its Nature and Justification (New York: Oxford,

2005).

8. See, for example, Sidney Bloch and Peter Reddaway, Soviet Psychiatric Abuse: The

Shadow over World Psychiatry (Boulder, CO: Westview Press, 1984), esp. ch. 1.

9. Paul S. Appelbaum, “The Parable of the Forensic Physician: Ethics and the Problem of

Doing Harm,” International Journal of Law and Psychiatry 13 (1990): 249–259, esp. 250–251.

10. Edmund Pellegrino and David Thomasma, For the Patient’s Good: The Restoration of

Beneficence in Health Care (New York: Oxford University Press, 1988); Pellegrino, “The

Four Principles and the Doctor-Patient Relationship: The Need for a Better Linkage,” in

Raanan Gillon, ed. Principles of Health Care Ethics (Chichester, England: John Wiley &

Sons, 1994).

11. Peter Singer, “Living High and Letting Die,” Philosophy and Phenomenological

Research 59 (1999): 183–187. Peter Singer, Practical Ethics, 2nd edition (Cambridge:

Cambridge University Press, 1993); Shelly Kagan, The Limits of Morality (Oxford:


Clarendon Press, 1989); Richard W. Miller, “Beneficence, Duty, and Distance,” Philosophy

& Public Affairs 32 (2004): 357–383.

12. For diverse accounts of justice in biomedical ethics, see Norman Daniels, Just Health

Care (New York: Cambridge University Press, 1985); Daniels, Just Health (New York:

Cambridge University Press, 2006); Madison Powers and Ruth Faden, Social Justice: The

Moral Foundations of Public Health and Health Policy (New York: Oxford University Press,

2006); Allen Buchanan, “Health-Care Delivery and Resource Allocation,” inMedical Ethics,

ed. Robert Veatch, 2nd edition (Boston: Jones and Bartlett Publishers, 1997); Allen

Buchanan, Dan Brock, Norman Daniels, and Daniel Wikler, From Chance to Choice:

Genetics and Justice (New York: Cambridge University Press, 2000).

13. Sissela Bok, Common Values (Columbia, MO: University of Missouri Press, 1995),

13–23, 50–59.

14. G. J. Warnock, The Object of Morality (London: Methuen & Co., 1971), esp. 15–26;

John Mackie, Ethics: Inventing Right and Wrong (London: Penguin, 1977), 107ff.

15. Henry S. Richardson, “Specifying Norms as a Way to Resolve Concrete Ethical

Problems,” Philosophy and Public Affairs 19 (Fall 1990): 279–310; Richardson, “Specifying,

Balancing, and Interpreting Bioethical Principles,” Journal of Medicine and Philosophy 25

(2000): 285–307; David DeGrazia, “Moving Forward in Bioethical Theory: Theories, Cases,

and Specified Principlism,” Journal of Medicine and Philosophy 17 (1992): 511–539; and

DeGrazia and Tom L. Beauchamp, “Philosophical Foundations and Philosophical

Methods,” in Methods of Bioethics, ed. D. Sulmasy and J. Sugarman (Washington, DC:

Georgetown University Press, 2001), esp. 33–36.

16. Richardson, “Specifying, Balancing, and Interpreting Bioethical Principles,” 289.

17. Norman Daniels, “Wide Reflective Equilibrium and Theory Acceptance in Ethics,”

Journal of Philosophy 76 (1979): 256–282; Daniels, “Wide Reflective Equilibrium in

Practice,” in L.W. Sumner and J. Boyle, Philosophical Perspectives on Bioethics (Toronto:

University of Toronto Press, 1996).

18. DeGrazia and Beauchamp, “Philosophical Foundations and Philosophical

Methods”; and David DeGrazia, “Common Morality, Coherence, and the Principles of

Biomedical Ethics,” Kennedy Institute of Ethics Journal 13 (2003): 219–230.

19. See, for example, B. Gert, C. M. Culver, and K. D. Clouser, Bioethics: A Return to

Fundamentals (New York: Oxford University Press, 1997), ch. 4; John H. Evans,

“A Sociological Account of the Growth of Principlism,” Hastings Center Report 30

(Sept.–Oct. 2000): 31–38; Earl Winkler, “Moral Philosophy and Bioethics: Contextualism

versus the Paradigm Theory,” in L. W. Sumner and J. Boyle, eds. Philosophical Perspectives

on Bioethics; John D. Arras, “Principles and Particularity: The Role of Cases in Bioethics,”

Indiana Law Journal 69 (1994): 983–1014; and Carson Strong, “Specified Principlism,”

Journal of Medicine and Philosophy 25 (2000): 285–307.


4

INFORMED CONSENT : ITS HISTORY

AND MEANING

The practice of obtaining informed consent has its history in medicine and

biomedical research, where the disclosure of information and the withholding

of information are aspects of the daily encounters between patient and physician,

as well as subject and investigator. Although discussions of disclosure and justified

nondisclosure have played a role in the history of medical ethics, the term

informed consent emerged only in the 1950s, and discussions of the concept as it

is used today began only around 1972. Concomitantly, a revolution was occurring

in standards of appropriate patient–physician interaction. Medical ethics moved

from a narrow focus on the physician’s or researcher’s obligation to disclose

information to the quality of a patient’s or subject’s understanding of information

and right to authorize or refuse a biomedical intervention.

HISTORICAL BACKGROUND

Many writers have proposed that managing medical information in encounters

with patients is a basic moral responsibility of physicians. Pioneering ventures

are found in classic documents in the history of medicine, such as the

Hippocratic writings (fifth to fourth century, BC), Percival’s Medical Ethics

(1803), and the first Code of Ethics (1846–1847) of the American Medical

Association, as well as in the historically significant didactic writings on medical

ethics in the eighteenth and nineteenth centuries, sometimes referred to as the

“learned” tradition, comprising discursive study of medical ethics through

treatises and books.

These codes and writings present a disappointing history from the perspec-

tive of the right to give informed consent. The Hippocratic writings did not hint

at obligations of veracity or disclosure, and throughout the ancient, medieval,

and early modern periods, medical ethics developed predominantly within the

profession of medicine. With few exceptions, no serious consideration was

given to issues of either consent or self-determination by patients and research

subjects. The central concern was how to make disclosures without harming

patients by revealing their condition too abruptly and starkly. The emphasis on

the principle “first, do no harm” promoted the idea that a health care profes-

sional is obligated not to make disclosures because to do so would be to risk a

harmful outcome.

The Eighteenth Century

Benjamin Rush and John Gregory have sometimes been acknowledged for

their enlightenment-inspired views about disclosure and public education in

the eighteenth century. However, neither was advocating informed consent.

They wanted patients to be sufficiently educated so that they could under-

stand their physician’s recommendations and therefore be motivated to

comply. They were not optimistic that patients would form their own opi-

nions and make appropriate medical choices. For example, Rush advised

physicians to “yield to them [patients] in matters of little consequence, but

maintain an inflexible Authority over them in matters that are essential to life”

(Rush, 1786, 323). Gregory underscored that the physician must be keenly aware of

the harm that untimely revelations might cause. There is no assertion of the

importance of respecting rights of self-determination for patients or of obtaining

consent for any purpose other than a medically good outcome (Gregory, 1772).

Gregory and Rush appreciated the value of information and dialogue from the

patient’s point of view, but the idea of informed consent was not foreshadowed

in their writings.

The Nineteenth Century

Thomas Percival’s historic Medical Ethics (1803) stands in this same tradition. It

makes no more mention of consent solicitation and respect for decision making

by patients than had previous codes and treatises. Percival did struggle with the

I N F O R M E D C O N S E N T : I T S H I S T O R Y A N D M E A N I N G 51

issue of truth telling, but he held that the patient’s right to the truth must yield in

the face of an obligation to benefit the patient in cases of conflict—a benevolent

deception. Percival maintained that

to a patient . . .who makes inquiries which, if faithfully answered, might


truth. His right to it is suspended, and even annihilated; because, its

beneficial nature being reversed, it would be deeply injurious to himself, to

his family, and to the public. And he has the strongest claim, from the trust

reposed in his physician, as well as from the common principles of humanity,

to be guarded against whatever would be detrimental to him. . . . The only

point at issue is, whether the practitioner shall sacrifice that delicate sense of

veracity, which is so ornamental to, and indeed forms a characteristic

excellence of the virtuous man, to this claim of professional justice and

social duty (Percival, 1803, 165–166).

Percival was struggling against the arguments of his friend, the Rev. Thomas

Gisborne, who opposed practices of giving false assertions intended to raise

patients’ hopes and lying for the patient’s benefit: “The physician . . . is invariably

bound never to represent the uncertainty or danger as less than he actually

believes it to be” (Gisborne, 1794, 401). From Percival’s perspective, the physician

does not lie or act improperly in beneficent acts of deception and falsehood as

long as the objective is to give hope to the dejected or sick patient.

The American Medical Association (AMA) accepted virtually without modifi-

cation the Percival paradigm in its 1847 Code (American Medical Association,

1847, 94). Many passages in Percival appear almost verbatim in this Code (AMA,

1847, “Code of Medical Ethics,” see esp. Ch. I, Art. I, sect. 4). This Code, as well as

most codes of medical ethics before and after, does not include rules of veracity.

For more than a century thereafter, American and Britishmedical ethics remained

fundamentally under Percival’s vision.

There was, however, a notable nineteenth-century exception to the consensus

that surrounded Percival’s recommendations. Connecticut physician Worthington

Hooker was the first champion of the rights of patients to information, in

opposition to the model of benevolent deception that had reigned from

Hippocrates to the AMA. He and Harvard Professor of Medicine Richard

Clarke Cabot may have been the only physician champions of this model prior

to the second half of the twentieth century. Moreover, there may never have been

a figure who swam, in regard to truth telling, so against the stream of indigenous

medical tradition as Hooker.

Hooker’s arguments are novel and ingenious, but do not amount to a

recommendation of informed consent. Hooker was concerned with “the


general effect of deception” on society and on medical institutions. He thought

the effect disastrous. But no more in Hooker than in the AMA Code is there a

recommendation to obtain the permission of patients or to respect autonomy

for the sake of autonomy. Hooker’s concerns were with expediency in dis-

closure and truth telling rather than with the promotion of autonomous

decision making or informed consent. The idea that patients should be enabled

to understand their situation so that they are able to participate with physi-

cians in decisions about medical treatment was an idea whose time had yet to

come (Hooker, 1859).

Although the nineteenth century saw no hint of a rule or practice of informed

consent in clinical medicine, consent procedures were not entirely absent.

Evidence exists in surgery records that consent seeking and rudimentary rules

for obtaining consent existed since at least the middle of the nineteenth century

(Pernick, 1982). However, the consents obtained do not appear to have been

meaningful by standards of informed consent, as they had little to do with the

patient’s right to decide after being informed. Practices of obtaining consent in

surgery prior to the 1950s were pragmatic responses to a combination of concerns

about medical reputation, malpractice suits, and practicality in medical institu-

tions. It is physically difficult and interpersonally awkward to perform surgery on

a patient without obtaining the patient’s permission. Such practices of obtaining

permission, however, did not constitute practices of obtaining informed consent,

although they did provide a modest nineteenth-century grounding for this

twentieth-century concept.

The situation is similar in research involving human subjects. Little evidence

exists that, until recently, requirements of informed consent had a significant

hold on the practice of investigators. In the nineteenth century, for example, it

was common for research to be conducted on slaves and servants without

consent on the part of the subject. By contrast, at the turn of the century,

American Army surgeon Walter Reed’s yellow fever experiments involved

formal procedures for obtaining the consent of potential subjects. Although

deficient by contemporary standards of disclosure and consent, these proce-

dures recognized the right of the individual to refuse or authorize participation

in the research. The extent to which this principle became ingrained in the

ethics of research by the mid-twentieth century remains a matter of historical

controversy.

Although it has often been reported that the obtaining of informed and

voluntary consent was essential to the ethics of research and was commonplace

in biomedical investigation, it is unclear that consent seeking on the part of

investigators was standard practice. Anecdotal evidence suggests that biomedical

research often proceeded without adequate consent at least into the 1960s (Faden &

Beauchamp, 1986).


Early Twentieth-Century Legal History

The legal history of disclosure obligations and rights of self-determination for

patients evolved gradually. In the doctrine of legal precedent, each decision,

relying on earlier court opinions, joins a chain of authority that incorporates

the relevant language and reasoning from the cited cases. In this way, a few early

consent cases built on each other to produce a legal doctrine. The best known and

ultimately the most influential of these early cases is Schloendorff v. Society of New

York Hospitals (1914). Schloendorff used rights of “self-determination” to justify

imposing an obligation to obtain a patient’s consent. Subsequent cases that

followed and relied upon Schloendorff implicitly adopted its reasoning. In this

way, “self-determination” came to be the primary justification for legal require-

ments that consent be obtained from patients.

In the early twentieth century, the behavior of physicians was often egregious

in the neglect of consent, and courts did not shrink from using ringing language

and sweeping principles to denounce it. The same language was then applied as

precedent in later cases in which physicians’ behavior was less outrageous. As the

informed consent doctrine developed and problems grew more subtle, the law

could have turned away from the language of self-determination, but instead

increasingly relied on this rationale as its fundamental premise. The language in

the early cases suggests that rights of freedom from bodily invasion contain rights

of medical decision making by patients.

The 1950s and 1960s

During the 1950s and 1960s, the traditional duty to obtain consent evolved into

a new, explicit duty to disclose certain forms of information and then to obtain

consent. This development needed a new term, and so “informed” was tacked

onto “consent,” creating the expression “informed consent,” in the landmark

decision in Salgo v. Leland Stanford, Jr. University Board of Trustees (1957). The

Salgo court suggested, without accompanying analysis, that the duty to disclose

the risks and alternatives of treatment was not a new duty, but a logical

extension of the already established duty to disclose the treatment’s nature

and consequences. Nonetheless, Salgo clearly introduced new elements into the

law. The Salgo court was not interested merely in whether a recognizable

consent had been given to the proposed procedures. The Salgo court latched

tenaciously onto the problem of whether the consent had been adequately

informed. The court thus created the language and the substance of informed

consent by invoking the same right of self-determination that had heretofore

applied to a less robust consent requirement.


Shortly thereafter, two opinions by the Kansas Supreme Court in the case of

Natanson v. Kline (1960) pioneered the use of negligence in informed consent

cases, rather than using battery. The court established the duty of disclosure as the

obligation “to disclose and explain to the patient in language as simple as

necessary the nature of the ailment, the nature of the proposed treatment, the

probability of success or of alternatives, and perhaps the risks of unfortunate

results and unforeseen conditions within the body.” Thus, the Natanson court

required essentially the same extensive disclosure—of the nature, consequences,

risks, and alternatives of a proposed procedure—as had Salgo.

Not surprisingly, the number of articles in the medical literature on issues of

consent increased substantially following this and other precedent legal cases.

Typically written by lawyers, these articles functioned to alert physicians both to

“informed consent” as a new legal development and to potential malpractice risk.

How physicians reacted to these legal developments in the 1950s and 1960s is not

well documented, but some empirical studies of informed consent in clinical

medicine provide insights. A study done in the early to mid-1960s, conducted by

the lawyer–surgeon team of Nathan Hershey and Stanley H. Bushkoff, indicates

that a preoperative consent form was not yet a ubiquitous feature of the practice of

surgery. As part of their study, Hershey and Bushkoff required that cooperating

surgeons complete a “fill-in-the-blank” consent form for each of their patients.

However, surgeons at several hospitals objected to the study because they were

not currently using any kind of consent form for surgery. In the end, only 10

surgeons, representing but 3 hospitals, participated in the study. Together, these

surgeons provided data on “informed consent” in 256 surgical cases. From this

limited sample, Hershey and Bushkoff inferred that a consistent standard of

disclosure existed among the surgeons studied and that this standard included

a description of the operative procedure and its attendant risks and consequences

(Hershey & Bushkoff, 1969).

Indifference to consent procedures seems to have begun to change by the late

1960s, when most physicians appear to have recognized some moral and legal

duty to obtain consent for certain procedures and to provide some kind of

disclosure. There is also evidence, however, that physicians’ views about proper

consent practices even in the late 1960s differed markedly from the consensus of

opinion and convention that would begin to be fixed only 10 years later. For

example, in one study, half of the physicians surveyed thought it medically

proper, and 30% ethically proper, for a physician to perform a mastectomy

with no authorization from the patient other than her signature on the blanket

consent form required for hospital admission. Also, more than half of the

physicians thought that it was ethically appropriate for a physician not to tell a

cancer patient that she had been enrolled in a double-blind clinical trial of an

experimental anticancer drug (Hagman, 1970).


An explosion of commentary on informed consent emerged in the medical

literature of the early 1970s, but much of this commentary was negative:

Physicians saw the demands of informed consent as impossible to fulfill and, at

least in some cases, inconsistent with good patient care. Nonetheless, empirical

studies conducted at the time suggest that there was at least enough documentable

consent seeking in such areas as surgery, organ donation, and angiography to

warrant empirical investigation. During this period, the procedure-specific consent

form was gaining acceptance, although it was not yet universally in use. Whether in

the 1960s physicians generally regarded informed consent as a legal nuisance or as

an important moral problem is unclear (Faden & Beauchamp, 1986).

The histories of informed consent in research and in clinicalmedicine were at this

time developing largely as separate pieces in a larger mosaic of biomedical ethics.

These pieces have never been well integrated even when they developed simulta-

neously. Research ethics prior toWorldWar II was no more influential on research

practices than the parallel history of clinical medicine was on clinical practices. But

events that unquestionably influenced thought about informed consent occurred at

the Nuremberg trials. The Nuremberg Military Tribunals unambiguously con-

demned the sinister political motivation of Nazi experiments. A list of 10 principles

constituted the Nuremberg Code. Principle One of the code states, without quali-

fication, that the primary consideration in research is the subject’s voluntary

consent, which is “absolutely essential” (United States v. Karl Brandt, 1947).

The Nuremberg Code served as a model for some professional and governmental

Codes formulated in the 1950s and 1960s. During this period, several additional

incidents involving consent violations moved the discussion of Post-Nuremberg

problems into the public arena. Thus began a rich and complex interplay of

influences on research ethics: scholarly publications, journalism, public outrage,

legislation, and case law. In the United States, one of the first incidents to achieve

notoriety in research ethics involved a study conducted at the Jewish Chronic

Disease Hospital (JCDH) in Brooklyn, New York. In July 1963, Dr. Chester

Southam of the Sloan-Kettering Institute for Cancer Research persuaded the

hospital’s medical director, Emmanuel E. Mandel, to permit research involving

injection of a suspension of foreign, live cancer cells into 22 patients at the JCDH.

The objective was to discover whether a decline in the body’s capacity to reject

cancer transplants was caused by their cancer or by debilitation. Patients without

cancer were needed to supply the answer. Southam had convinced Mandel that

although the research was nontherapeutic, such research was routinely done

without consent. Some patients were informed orally that they were involved in

an experiment, but it was not disclosed that they were being given injections of

cancer cells. No written consent was attempted, and some subjects were incompe-

tent to give informed consent. In 1966, the Board of Regents of the State University

of New York censuredDrs. Southam andMandel for their role in the research. They


were found guilty of fraud, deceit, and unprofessional conduct (Hyman v. Jewish

Chronic Disease Hospital, 1965).

Another major controversy about the ethics of research in the United States

developed at Willowbrook State School, an institution for “mentally defective”

children on Staten Island, New York. Beginning in 1956, Saul Krugman and his

associates began a series of experiments to develop an effective prophylactic agent

for infectious hepatitis. They deliberately infected newly admitted patients with

isolated strains of the virus based on parental consents obtained under controversial

circumstances that may have been manipulative. The issues in the Willowbrook

case are more complex than those in the Jewish Chronic Disease Hospital case, and

still today there are those who defend, at least in part, the ethics of these experi-

ments. Krugman’s research unit was eventually closed, but closure on the debate

about the ethics of the studies conducted in the unit was never achieved

(Proceedings of the Symposium on Ethical Issues in Human Experimentation, 1972).

The most notorious case of prolonged and knowing violation of subjects’ rights

in the United States was a Public Health Service (PHS) study initiated in the early

1930s. Originally designed as one of the first syphilis control demonstrations in the

United States, the stated purpose of the Tuskegee Study, as it is now called, was to

compare the health and longevity of an untreated syphilitic population with a

nonsyphilitic but otherwise similar population. These subjects, all African

American males, knew neither the name nor the nature of their disease. Their

participation in a nontherapeutic experiment also went undisclosed. They were

informed only that they were receiving free treatment for “bad blood,” a term local

African Americans associated with a host of unrelated ailments, but which the white

physicians allegedly assumed was a local euphemism for syphilis (Jones, 1993).

It was remarkable that, although this study was reviewed several times between

1932 and 1970 by PHS officials and medical societies, and reported in 13 articles in

prestigious medical and public health journals, it continued uninterrupted and

without serious challenge. It was not until 1972 that the Department of Health,

Education, and Welfare (DHEW) appointed an ad hoc advisory panel to review

the study and the department’s policies and procedures for the protection of

human subjects. The panel found that neither the DHEW nor any other govern-

ment agency had a uniform or adequate policy for reviewing experimental

procedures or securing subjects’ consents.

The 1970s and 1980s

These events do not, of course, explain why informed consent became the focus of

so much attention in both case law and biomedical ethics. Many hypotheses can

be invoked to explain these developments. The most likely explanation is that law,


ethics, and medicine were all affected by issues and concerns in the wider society

regarding individual liberties and social equality. These issues were made dra-

matic by an increasingly technological, powerful, and impersonal medical care.

The issues raised by civil rights, women’s rights, the consumer movement, and the

rights of prisoners and the mentally ill often included health care components and

helped reinforce public acceptance of rights applied to health care. Informed

consent was swept along with this body of social concerns, which propelled the

new bioethics throughout the 1970s.

Three 1972 court decisions stand as informed consent landmarks: Canterbury

v. Spence, Cobbs v. Grant, and Wilkinson v. Vesey. Canterbury had a particularly

massive influence in demanding a more patient-oriented standard of disclosure.

In Canterbury, surgery on the patient’s back and a subsequent accident in the

hospital led to further injuries and unexpected paralysis, the possibility of which

had not yet been disclosed. Judge Spottswood Robinson’s opinion focuses on the

needs of the reasonable person and the right to self-determination. As for suffi-

ciency of information, the court holds: “The patient’s right of self-decision shapes

the boundaries of the duty to reveal. That right can be effectively exercised only if

the patient possesses enough information to enable an intelligent choice.”

Among the most important publications in the medical literature to appear

during this period was a statement by the Judicial Council of the American

Medical Association in 1981. For the first time, the AMA recognized informed

consent as “a basic social policy” necessary to enable patients to make their own

choices, even if their physician disagrees. The AMA’s statement is a testament to

the impact of the law of informed consent on medical ethics. The AMA’s position

roughly followed the language of Canterbury v. Spence.

The President’s Commission for the Study of Ethical Problems inMedicine and

Biomedical and Behavioral Research provides further evidence about the status

informed consent had achieved in the 1970s and early 1980s. The President’s

Commission was first convened in January 1980, with informed consent as a

main item on its agenda. In 1982, it produced one three-volume report that dealt

directly with informed consent: Making Health Care Decisions: The Ethical and

Legal Implications of Informed Consent in the Patient-Practitioner Relationship. The

President’s Commission argued that although informed consent had emerged

primarily from a history in law, its requirements are essentially moral and policy

oriented. It held that informed consent is based on the principle that competent

persons are entitled to make their own decisions from their own values and goals,

but that the context of informed consent and any claim of “valid consent” must

derive from active, shared decision making. The principle of self-determination

was described as the “bedrock” of the President’s Commission’s viewpoint.

The 1980s saw the publication of several books devoted to the subject of

informed consent, as well as hundreds of journal articles and the passage of


procedure-specific, informed consent laws and regulations. These events give

testimony to the importance of informed consent in moral and legal thinking

about medicine in the United States. By themselves, however, they tell us little

about physicians’ or researchers’ actual consent practices or opinions, or about

how informed consent was viewed or experienced by patients and subjects.

As might be expected, the empirical evidence on this subject during the 1980s is

mixed, although it is clear that procedures of informed consent had taken a firm

hold in some parts of medical practice. For example, routine practice encouraged

the obtaining of signatures on consent forms and the disclosing of information

about alternative treatments, risks, and benefits. The best data on this subject are the

findings of a national survey conducted for the President’s Commission by Louis

Harris and Associates in 1982. Almost all of the physicians surveyed indicated that

they obtained written consent from their patients before inpatient surgery or the

administration of general anesthesia. At least 85% said they usually obtained some

kind of consent—written or oral—for minor office surgery, setting of fractures,

local anesthesia, invasive diagnostic procedures, and radiation therapy. Only blood

tests and prescriptions appear to have proceeded frequently without patient con-

sent, although about half of the physicians reported obtaining oral consent.

The overall impression conveyed by this survey is that the explosion of interest

in informed consent in the 1970s had a powerful impact on medical practice.

However, evidence from the Harris survey and other sources raises questions

about the quality and meaningfulness of this consent-related activity. The over-

whelming impression from the empirical literature and from reported clinical

experience is that the actual process of soliciting informed consent often fell short

of a serious show of respect for the decisional authority of patients. As the authors

of one empirical study of physician–patient interactions put it, “despite the

doctrine of informed consent, it is the physician, and not the patient, who, in

effect, makes the treatment decision” (Siminoff & Fetting, 1991, 817).

The history of informed consent, then, indicates that medicine has undergone

widespread changes under the influence of legal and moral requirements of

informed consent, but it also reminds us that informed consent is an evolving

process, not a set of events whose history has passed.

THE CONCEPT AND ELEMENTS OF INFORMED CONSENT

Informed consent began to play a central role in clinical and research ethics when

problems of the autonomy of subjects gradually grew more insistent in twentieth-

century practice and research, and when the idea of respecting autonomy gained

equal recognition as a form of justification for protecting against risk. The

practice of obtaining consent is a social phenomenon, and no analysis of the


concept of informed consent will succeed if it ignores the contexts in which

informed consent arose. Considerable vagueness has attended the term informed

consent in these contexts, leaving a need to sharpen the concept.

Presumptions About the Concept

The claim that something is an informed consent or that an informed consent has

been obtained cannot always be taken at face value. Before we can confidently

infer that what appears to have been or was called an “informed consent” is a bona

fide instance of informed consent, we need to know what to look for. This inquiry

requires criteria of what will qualify for the label “informed consent.”

If overdemanding criteria such as “full disclosure and complete understanding”

are adopted, an informed consent becomes impossible to obtain. Conversely, if

underdemanding criteria such as “the patient signed the form” are used, an

informed consent becomes too easy to obtain and the term loses all moral

significance. Many interactions between a physician and a patient or an investi-

gator and a subject that have been called informed consents have been so labeled

only because they rest on underdemanding criteria; they are inappropriately

referred to as informed consents. For example, a physician’s truthful disclosure

to a patient has often been declared the essence of informed consent, as if a

patient’s silence following disclosure could add up to an informed consent. The

existence of such inadequate understandings of informed consent can be

explained in part by empirical information about physicians’ beliefs concerning

informed consent.

Contemporary Presumptions in Medicine

Some data about the meaning of informed consent are found in the aforemen-

tioned survey of physicians conducted by Louis Harris (Harris, in President’s

Commission, 1982, 302), which asked physicians, “What does the term informed

consent mean to you?” In their answers, only 26% of physicians indicated that

informed consent had anything to do with a patient giving permission, con-

senting, or agreeing to treatment; only 9% indicated that it involved the patient

making a choice or stating a preference about his or her treatment. Similar results

have been found in recent surveys of Japanese physicians’ beliefs about informed

consent (Hattori, 1991; Kai, 1993; Mizushima, 1990; Takahashi, 1990).

Like lawyers and courts, the overwhelming majority of these doctors appeared

to recognize only disclosure as the criterion of “informed consent.” That is, they

view informed consent as explaining to a patient the nature of his or her medical


condition together with a recommended treatment plan. But if physicians regard

informed consent as nothing more than an event of conveying information to

patients, rather than a process of discussion and obtaining permission from the

patient, then claims that they regularly “obtain consents” from their patients

before initiating medical procedures are vague and unreliable unless we know in

some detail about the procedures used.

Matters may be worse than they appear: Perhaps all these physicians under-

stand by “informed consent” is that the patient’s signature has been obtained, or

perhaps they mean only that some kind of disclosure has been made. This

interpretation fits with the results of several studies of informed consent that

have failed to find any sizeable evidence of “informed consent” in clinical med-

icine and that have found little evidence that the consents being obtained are

meaningful exercises of informed choice by patients (Quaid, 1990, 249–259;

Scherer & Reppucci, 1988, 123–141; Siminoff & Fetting, 1991, 813–818; Siminoff,

Fetting, & Abeloff, 1989, 1192–1200).

The Authority of Oaths, Codes, and Treatises

Similar problems exist regarding what can be reasonably inferred from oaths,

prayers, Codes of ethics, published lectures, and general pamphlets and treatises

on medical conduct, usually written by individual physicians or medical societies

for their colleagues. In the absence of more direct data about actual consent

practices, these documents have been relied on heavily in writings on informed

consent as sources that provide information about the history of informed consent

and related matters of clinical medical ethics. However, it is often difficult to

determine whether the statements that appear in these documents are primarily

exhortatory, descriptive, or self-protective. Some writings describe, for educational

purposes, conduct that was in accordance with prevailing professional standards.

Other documents aim at reforming professional conduct by prescribing what

should be established practice. Still others seem constructed to protect the physician

from suspicions of misconduct or from legal liability.

The Elements of Informed Consent

Legal, philosophical, regulatory, medical, and psychological literatures have often

tried to define or analyze informed consent in terms of its “elements.” The following

elements have been widely mentioned as fundamental to the concept: (1) disclo-

sure, (2) comprehension, (3) voluntariness, (4) competence, and (5) consent

(Levine, 1978, 3–9; Meisel & Roth, 1981; National Commission, 1978, 10). The


postulate is that a person gives an informed consent to an intervention if and only

if the person is competent to act, receives a thorough disclosure about the

procedure, comprehends the disclosed information, acts voluntarily, and

consents.

This definition is attractive because of its consistency with standard usage

of the term informed consent in medicine and law. However, medical conven-

tion and malpractice law tend to distort the meaning of informed consent in

ways that need correction. Analyses using the aforementioned five elements,

as well as conventional usage in law and medicine, are best suited for catalog-

ing the analytical parts of informed consent and for delineating moral and

legal requirements of informed consent, not for conceptually analyzing the

meaning of “informed consent.” Neither requirements nor parts amount to a

definition.

To take but one instance of the potential bias at work in this form of

definition, the U.S. Supreme Court addressed the definition of informed con-

sent in Planned Parenthood of Central Missouri v. Danforth as follows: “One

might well wonder . . .what ‘informed consent’ of a patient is. . . . We are

content to accept, as the meaning [of informed consent], the giving of informa-

tion to the patient as to just what would be done and as to its consequences”

(1976, 67, n. 8). The exclusive element of informed consent here is disclosure,

which recalls the assumptions made by physicians in the Harris poll. However,

nothing about an informed consent requires disclosure as part of its meaning.

A patient or subject already knowledgeable about a proposed intervention

could give a thorough informed consent without having received a disclosure

from a second party. Similarly, other conditions in the aforementioned list of

conditions are not necessary. For example, persons who are legally incompetent

(see element 4) sometimes give informed consents, and in some instances even

psychologically incompetent persons (also often the referent of element 4) may

be able to consent meaningfully to or refuse a particular intervention. These

norms delineate an obligation to make disclosures so that a consent can be

informed, rather than a meaning of informed consent. Even all five of these

elements merged as a set do not satisfactorily capture the meaning of “informed

consent.”

The following seven categories express the analytical components of informed

consent more adequately than the aforementioned five categories, but this seven-

fold list still does not adequately express the meaning of “informed consent”

(Beauchamp & Childress, 1994):

I. Threshold elements (preconditions)

1. Competence (to understand and decide)

2. Voluntariness (in deciding)


II. Information elements

3. Disclosure (of material information)

4. Recommendation (of a plan)

5. Understanding (of elements 3 and 4)

III. Consent elements

6. Decision (in favor of a plan)

7. Authorization (of the chosen plan)

The language of “material information” in element 3 is pivotal for an adequate

analysis of the elements of disclosure (element 3) and understanding (element 5).

Critics of legal requirements of informed consent have often held that procedures

sometimes have so many risks and benefits that they cannot be disclosed and

explained, but “material risks” are simply the risks a reasonable patient needs to

understand in order to decide among the alternatives. Only these risks and

benefits need to be disclosed and understood.

Informed consent requirements can be constructed to correspond to each of the

aforementioned elements. That is, specific disclosure requirements, comprehension

requirements, noninfluence requirements, competence requirements, authoriza-

tion requirements, and the like can be fashioned. These requirements would specify

the conditions that must be satisfied for a consent to be valid, but they may not

provide an adequate analysis of the meaning of “informed consent.”

Two Meanings of “Informed Consent”

The question “What is an informed consent?” is complicated because at least two

common, entrenched, and irreducibly differentmeanings of “informedconsent”have

been at work in its history. That is, the term is analyzable in different ways because

different conceptions of informed consent have emerged. In one sense, an informed

consent is an autonomous authorization by individual patients or subjects. In the

second sense, informed consent is analyzable in termsof institutional andpolicy rules of

consent that collectively form the social practice of informed consent in institutional

contexts (Faden & Beauchamp 1986, 276–287).

In the first meaning, an autonomous authorization requires more than merely

acquiescing in, yielding to, or complying with an arrangement or a proposal made

by a physician or investigator. A person gives an informed consent in this first

sense if and only if the person, with substantial understanding and in substantial

absence of control by others, intentionally authorizes a health professional to do

something. A person who intentionally refuses to authorize an intervention, but

otherwise satisfies these conditions, gives an informed refusal. This first sense


derives from philosophical premises that informed consent is fundamentally a

matter of protecting and enabling autonomous or self-determining choice.

In the second meaning, “informed consent” refers only to a legally or institu-

tionally effective approval given by a patient or subject. An approval is therefore

“effective” or “valid” if it conforms to the rules that govern specific institutions,

whatever the operative rules may be. In this sense, unlike the first, conditions and

requirements of informed consent are relative to a social and institutional context

and need not be autonomous authorizations. This meaning is driven by demands

in the legal and health care systems for a generally applicable and efficient consent

mechanism by which responsibilities and violations can be readily and fairly

assessed.

Under these two contrasting understandings of “informed consent,” a patient

or subject can give an informed consent in the first sense, but not in the second

sense, and vice versa. For example, if the person consenting is a minor and

therefore not of legal age, he or she cannot give an effective or valid consent

under the prevailing institutional rules; a consent is invalid even if the minor gives

the consent autonomously and responsibly. (“Mature minor” laws do sometimes

make an exception and give minors the right to authorize medical treatments in a

limited range of circumstances.)

“Informed consent” in the second sense, as institutional consent, has until very

recently constituted the mainstream conception in the regulatory rules of federal

agencies as well as in health care institutions. The documents governing consent

in these contexts derive from a conception of what the rules must be in order to

promote effective authorizations in these institutions, but the rules were only

rarely premised on a conception of autonomous authorization that had more

than a superficial quality. However, literature in bioethics has increasingly sug-

gested that any justifiable analysis of informed consent must be rooted in auton-

omous choice by patients and subjects.

In principle, although less clearly in practice, the conditions of informed

consent as autonomous authorization can function as model standards for fash-

ioning the institutional and policy requirements of informed consent—a model

of autonomous choice thus serving as the benchmark against which the moral

adequacy of prevailing rules and practices might be evaluated.

Autonomous Choice

It has often been said that the justification of requirements of informed consent

(in the first sense at least) is the principle of respect for autonomy. However, the

goal of ensuring that persons make “autonomous choices” has proved to be

difficult to implement. Historically, little more can be said beyond the fact that


a clear societal consensus has developed that there must be adequate protection of

patients’ and subjects’ decision-making rights, especially their autonomy rights.

We therefore need to examine the meaning of “autonomy” in addition to the

meaning of “informed consent.”

In the literature on informed consent, “autonomy” and “respect for autonomy”

are terms loosely associated with several ideas, such as privacy, voluntariness, self-

mastery, choosing freely, the freedom to choose, choosing one’s own moral posi-

tion, and accepting responsibility for one’s choices. Because of this conceptual

uncertainty, the concept of autonomy and its connection to informed consent need

careful analysis.

In moral philosophy, personal autonomy has come to refer to personal self-

governance: personal rule of the self by adequate understanding while remaining

free from controlling interferences by others and from personal limitations that

prevent choice. Many issues about consent concern failures to respect autonomy,

ranging from manipulative underdisclosure of pertinent information to nonre-

cognition of a refusal of medical interventions. To respect an autonomous agent is

to recognize with due appreciation the person’s capacities and perspective,

including his or her right to hold certain views and to take certain actions based

on personal values and beliefs. Accordingly, an informed consent based on an

autonomous authorization suggests a well-informed agent with the competence

to authorize or refuse authorization of a medical or research intervention.

It has sometimes been claimed that informed consent, so understood, has a

mythical quality, because true informed consent is never obtained under such a

high ideal. The idea is that most patients and subjects cannot comprehend enough

information or appreciate its relevance sufficiently to make decisions about

medical care or about participation in research. This objection, however, springs

from a misunderstanding of the nature and goals of informed consent, based in

part on unwarranted standards of full disclosure and full understanding. The ideal

of complete disclosure of all possibly relevant knowledge needs to be replaced by a

more acceptable account of how patients and subjects understand relevant infor-

mation. Merely because our actions fail to be fully informed, voluntary, or

autonomous is no indication that they are never adequately informed or auton-

omous. We would never autonomously sign contracts, have automobiles

repaired, file income-tax returns, and the like if this were the case.

This argument does not deny that some individuals have a knowledge base that

is so impoverished that autonomous decision making about alien or novel situa-

tions is exceedingly difficult, but even under difficult situations there may be no

reason to foreclose the possibility of making an adequate decision. Successful

communication of novel, alien, and specialized information to laypersons can be

accomplished by drawing analogies between the information and more ordinary

events with which the patient or subject is familiar. Similarly, professionals can


express probabilities in both numeric and nonnumeric terms while helping the

patient or subject to assign meanings to the probabilities through comparison

with more familiar risks and prior experiences.

THE LAW AND ITS LIMITS

The law of informed consent has been more influential as an authoritative set of

statements and source of reflection than any other body of thought on the subject.

“The doctrine of informed consent,” as it is sometimes called, is the legal doctrine;

and “informed consent” has often been treated as synonymous with this legal

doctrine, which derives from the common law and includes the entire body of law

dealing with the obligation to obtain consent. This legal vision is focused on

disclosure and on liability for injury. There are good reasons, as we shall now see,

why the law turns on such a narrow basis and also why it is ill-equipped to serve

beyond these boundaries.

Theory of Liability

The primary basis for the legal doctrine is tort law. A “tort” is a civil injury to

one’s person or property that is inflicted by another and that is measured in terms

of, and compensated by, money damages. This law imposes duties on members of

society, and one who fails to fulfill a legal duty is liable for compensation for the

misdeed (in the civil law). The theory of liability under which a case is tried

determines the duty that must be fulfilled. In recent informed consent cases,

negligence is the theory of liability almost uniformly applied. However, the

informed consent doctrine originally developed and flourished under battery,

which is a different theory of liability. Currently, no unified legal theory underlies

all informed consent cases.

Under battery theory, the defendant is held liable for any intended (i.e.,

not careless or accidental) action that results in physical contact for which

the plaintiff has given no permission. A defendant need not have an evil

intent, nor must injury result; the unpermitted contact is itself considered

wrongful. Under negligence theory, by contrast, unintentional, “careless”

action or omission is the source of liability. The carelessness occurs in

regard to some activity in which the defendant has a duty to take care or

to behave reasonably toward others, and an injury measurable in monetary

terms is caused by failure to discharge the duty (Berkey, 1969; Meisel, 1977;

Schloendorff, 1914).


The Duty of Disclosure

Two competing disclosure standards have evolved as attempts to resolve problems

regarding the nature and amount of the information that must be disclosed: the

professional practice standard and the reasonable person standard. A third stan-

dard, the subjective standard, has also been discussed in legal commentary. The

professional practice standard holds that both the range of the duty to disclose and

the criteria of adequate disclosure are properly determined by the customary

practices of a professional community. These practices establish the standards of

care for disclosure and care alike. The patient, subject, or reasonable person lacking

expert knowledge is considered unqualified to decide what should be disclosed.

Although the professional practice standard remains the primary standard in

informed consent law, it contains inadequacies and it may be doubted whether a

customary standard of disclosure exists for much of medical practice. A basic

problem is that negligent care might be perpetuated if relevant professionals

throughout the profession offer inferior information, and another doubtful pre-

mise is that physicians have sufficient expertise to be able to judge in many cases

which items of information their patients need. However, the principal objection

to this standard is its failure to promote decisional autonomy, the protection of

which is generally accepted as the primary function and moral justification of

informed consent requirements.

In contrast to the professional practice standard, the reasonable person stan-

dard focuses on the information a “reasonable person” needs to know about

procedures, risks, alternatives, and consequences. The legal test of an adequate

disclosure is the “materiality,” or significance, of information to the decision

making of the patient. Thus, the right to decide what information is material and

due is shifted away from the physician to the reasonable patient. The reasonable

person standard requires a physician to divulge any fact that is material to a

reasonable person’s decision, but no requirement exists to meet the unreasonable

demand of a patient.

This reasonable person standard is as vulnerable to criticism as the profes-

sional-based standard. It can be doubted whether the reasonable person standard

serves the interests of those patients who know little about either their informa-

tional needs or the medical system. The interpretation of the standard for clinical

practice is also difficult, because it specifies no precise duty for physicians. Both

the concept of material information and the central concept of the reasonable

person are left at an intuitive level. It can therefore be doubted whether the

reasonable person standard more adequately protects the patient’s right to

choose than does the professional practice standard.

These arguments are not intended to eliminate standards of disclosure. In the

absence of a disclosure initiated by a professional, patients or subjects often cannot


formulate their concerns and ask meaningful questions. A patient or subject needs

to understand what an informed professional judges to be of value for most

patients or subjects as material information, and what it means for consent to be

an authorization to proceed. The problem is not whether we need adequate

disclosures but whether a legal vehicle can be expected to provide an adequate

standard of disclosure for clinical practice and research settings. I shall return to

this issue under the discussion of “the quality of consent.”

The larger problem with these standards is that they will not be of major

assistance in formulating a conception of informed consent for clinical medicine

and research. Because courts are captivated by the context of after-the-fact

resolution of narrow and concrete questions of duty, responsibility, blame,

injury, and damages in specific cases, the law has no systematic way of affecting

contemporary medical practice other than by a somewhat muted threat of prose-

cution for legal wrongdoing.

For all of these reasons, the heart of issues about informed consent is not legal

but moral. Informed consent has less to do with the liability of professionals and

more to do with the understanding and autonomous choices of patients and

subjects.

THE QUALITY OF CONSENT

In discussing both autonomous choice and the limits of law, we have noted that

problems about the quality and adequacy of consent probably cannot be resolved

unless conventional disclosure rules are redirected toward the quality of under-

standing present in a “consent.” This approach focuses on the need for commu-

nication, dispensing with liability-oriented discussions about proper legal

standards of disclosure. The key to effective communication is to invite participa-

tion by patients or subjects in an exchange of information and dialogue. Asking

questions, eliciting the concerns and interests of the patient or subject, and

establishing a climate that encourages the patient or subject to ask questions

seem more important for medical ethics than requirements of disclosure in law.

Without a proper climate in the consent context, a request from a professional

that the patient or subject ask for information is as likely to result in silence as to

elicit the desired result of a meaningful informational exchange and consent.

Patients find it difficult to approach physicians with questions or concerns, and

even when they do not understand their physicians, many still do not ask ques-

tions. The extent to which this passive attitude characterizes research subjects is

less clear. Although still understudied, it is a good guess that relatively little

educating of this quality occurs at present in either clinical practice or research

in the United States.


VULNERABLE SUBJECTS AND COMPLIANT PATIENTS

Patients and subjects are entitled to expect that physicians and research inves-

tigators who request decisions will do so free of coercion and manipulation.

Much discussion about the morality of asking subjects and patients to consent

centers not on how informed the subjects are, but on how free they are. For

example, this topic dominated the deliberations of the National Commission

for the Protection of Human Subjects over the involvement of prisoners in

drug research. The National Commission raised the question of “whether

prisoners are, in the words of the Nuremberg Code, ‘so situated as to be able

to exercise free power of choice.’” The National Commission answered that

“although prisoners who participate in research affirm that they do so freely,

the conditions of social and economic deprivation in which they live compro-

mise their freedom. The Commission believes, therefore, that the appropriate

expression of respect consists in protection from exploitation” (National

Commission, 1976, 5–7).

The National Commission went on to recommend a ban on drug research with

prisoners, on grounds that in coercive institutions free choice would be too often

compromised. Six years later this argument continued: Robert Levine, who had

served on the staff of the National Commission, argued—as had a minority of

commissioners—that prisoners are actually better off, not worse off, by their

involvement in research and that exclusion from research was a deeper restriction

of free choice than allowing inclusion in the research (Levine, 1982, 6). Many were

dubious, however, that such abstract statements take account of the realities of

manipulation and coercion in the prison (Dubler, 1982, 10).

A similar controversy occurred in the 1980s when Dr. Mortimer Lipsett at the

National Institutes of Health (National Institute of Child Health and Human

Development) explored the question of whether phase I clinical trials of cancer

chemotherapies involve a special class of subjects deserving special protections.

He presented the problem as follows:

The President’s Commission for the Study of Ethical Problems in

Medicine and Biomedical and Behavioral Research developed and

extended the concept of the vulnerable subject in medical research.

Children, prisoners, and the mentally disabled were defined as

vulnerable because a variety of constraints and inducements effectively

removed their capacity to function autonomously. Similarly, patients with

advanced cancer are faced with inducements that may sway their judgment

of the risk-benefit ratio. Should such patients be treated as vulnerable

research subjects necessitating extraordinary supervision by third parties?

(Lipsett, 1982, 941–942)


Lipsett answered that every patient entering such a therapeutic trial is vulner-

able by virtue of the disease state and the unique opportunity to receive a

promising drug, but he maintained that the problem could be and was being

overcome by “painstaking consultation and preparation” involving families,

institutional review boards, third-party consultation, and the like. He concluded

that, as conducted, phase I clinical trials of “cancer chemotherapies are ethical and

necessary” (Lipsett, 1982, 941–942).

Some published responses to Lipsett were less sanguine: Alexander Capron,

who had been staff director of the President’s Commission, as well as Terrence

Ackerman and Carson Strong, argued that the notion of therapeutic intent in the

trials is easily subject to misunderstanding by patients, who may be misled by the

hope of a favorable effect, especially when the prospect for therapeutic efficacy is

as remote as it often is at the dosage level offered. They maintained that patients

should be given a realistic picture of how they are contributing to medical

knowledge and noted the dangers of “manipulating” subjects and of subjecting

them to “affective” factors that impair understanding and judgment. In general,

they challenged the view that present safeguards are sufficient to preclude exploi-

tation (Ackerman & Strong, 1983, 883; Capron, 1983, 882–883).

Underlying these discussions is a theoretically difficult and partially unresolved

set of problems about free choice, coercion, and manipulation. Certain forms of

withholding information, playing on emotion, or presenting constraints can rob

an autonomous person of the capacity of free choice through manipulation or

coercion. Deceptive and misleading statements limit freedom by restricting the

range of choice and by getting a person to do what the person otherwise would not

do (cf. Bok, 1992, 1118–1119). The National Commission was worried about coer-

cion in the case of prisoners; by contrast, those engaged in the discussion of phase I

trials and cancer chemotherapies were interested in manipulation.

A continuum of controlling influences is present in our daily lives, running

from coercion, which is at the most controlling end of the continuum (compare

the National Commission’s model of prisoners), to persuasion and education,

which are not controlling at all, even though they are influences. For an action to

be classified as either voluntary or nonvoluntary, cut-off points on the continuum

of control to noncontrol are required. To fix the point of voluntariness, only a

substantial satisfaction of the condition of noncontrol is needed. The line between

the substantial and the insubstantial can be fixed in light of specific objectives of

decision making. These lines will be influenced by moral views of when it is

appropriate to respect a person’s decision as substantially voluntary, but this

connection need not be pursued here.

Influence does not necessarily imply constraint, governance, force, or compul-

sion, although these concepts are essential to certain kinds of influence. Important

decisions are usually made in contexts replete with influences in the form of


competing claims and interests, social demands and expectations, and straight-

forward or devious attempts by others to bring about the outcome they desire.

Some of these influences are unavoidable, and some may be desirable. Not all of

them interfere with or deprive persons of autonomous belief and action, as when

patients are persuaded by sound reasons to do something.

Coercion that involves a threat of harm so severe that a person is unable to

resist acting to avoid it is always completely controlling. It entirely negates

freedom because it entirely controls action. Persuasion, by contrast, is the inten-

tional and successful attempt to induce a person, through appeals to reason, to

freely accept the beliefs, attitudes, values, intentions, or actions advocated by the

persuader. Like informing, persuading is entirely compatible with free choice.

The most sweeping and difficult area of influence is manipulation, a broad,

general category that runs from highly controlling to altogether noncontrolling

influences. The essence of manipulation is getting people to do what the

manipulator intends without resort to coercion or to reasoned argument. In a

paradigm case of manipulation in contexts of informed consent, information is

closely managed to bring about the “choice” by a patient that the physician

wants the patient to make, and the patient does not know what the manipula-

tive physician intends. Whether such uses of information necessarily compro-

mise or restrict free choice is an unresolved and untidy issue, but one plausible

view is that some manipulations (e.g., the use of rewards such as reduced

medical fees for being involved in research) are compatible with free choice,

whereas others (e.g., deceptive offers of hope of a cure where there is none) are

not compatible with free choice.

In informed consent contexts, the central question is not whether we are entirely

free of manipulative influences, but whether we are sufficiently free to remain

autonomous—free to perform our own actions—as opposed to controlled by the

actions of another. The thorniest of all problems about autonomy and manipula-

tion is not that of punishment and threat, but the effect of rewards and offers. This

category refers to the intentional use of offers of rewards to bring about a desired

response. For example, during the Tuskegee Syphilis experiments, various methods

were used to stimulate and sustain the interest of subjects in continued participa-

tion. They were offered free burial assistance and insurance, free transportation, and

a free stop in town on the return trip. They were also rewarded with free medicines

and free hot meals on the days of the examination. The deprived socioeconomic

condition of these subjects made them easily manipulable by those means.

This general range of problems is compounded further by what is sometimes

called the problem of “coercive situations,” but is better formulated as “con-

straining situations.” Most accounts of coercion require that coercion be inten-

tional by the coercer, but constraining situations suggest nonintentional “coercion”

where the person is controlled by the situation, not by the design of another person.


Sometimes people unintentionally make other persons feel “threatened,” and

sometimes situations of illness and economic necessity present “threats” of serious

harm that a person feels compelled to prevent at all costs. The earlier example of

using prisoners in experimentation is again applicable if we assume that a prisoner

is left without any viable alternative to participation in research because of the risks

of the alternatives in the circumstance or because of what may appear to be threats

presented by prison officials. As Alvin Bronstein, director of the National Prison

Project, once put the problem for informed consent, “You cannot create . . . [a

prison] institution in which informed consent without coercion is feasible”

(Bronstein, 1975, 130–131).

Beyond the prison, in circumstances of severe physical or health deprivation, a

personmight accept an offer or sign a contract that the person would refuse under

less stringent circumstances. Cancer patients provide good examples: The pro-

spect of death if an objectionable toxic drug is rejected seems to “coerce” a choice

of the drug no less than an intentional threat. The psychological effect on the

person forced to choose may be identical, and the person can appropriately say in

both cases, “There was no real choice; I would have been crazy to refuse.” But if, as

we usually believe, a contract signed under another person’s threat is invalid (and

the consent behind the signing an invalid consent), can we not say that a person

who agrees to drug experimentation in a constraining situation has made an

invalid contract (and given an invalid consent)?

It is a mistake to suppose that persons in such constraining situations cannot

act autonomously. A loss of alternatives cannot be equated with coercion. It is,

then, a confusion to move from a correct claim about a loss of options caused by

desperate circumstances to the conclusion that there has been a loss of autonomy

because of a constraining situation. Nevertheless, even if knowledgeable intent to

threaten is not present, one may feel just as forced to a choice and may just as

heartily wish to avoid it. Such perceptions are why the issues are often presented in

the language of “vulnerable subjects.” Their situation does make them vulnerable,

even desperately vulnerable, and may subject them to “control” by their emotions

and anxieties to an abnormal degree.

Despite the tense and pressured nature of such circumstances, most patients

have the ability to consent freely and will do so if properly managed. What may be

doubted is how often a free informed consent does in fact occur.

COMPETENCE TO CONSENT

I observed earlier that competence is commonly considered a necessary condition

of informed consent. In legal and policy contexts, reference to competent persons

is far more common than reference to autonomous persons. Competence


judgments function as a gatekeeping device for informed consent. That is, com-

petence judgments function to distinguish persons from whom consent should be

solicited from those from whom consent need not or should not be solicited. In

health care, competence judgments distinguish the class of individuals whose

autonomous decisions must be respected from those individuals whose decisions

need to be checked and perhaps overridden by a surrogate (Buchanan & Brock,

1989, 26–27; Faden & Beauchamp, 1986, 287–292).

Competence can be either a factual or a categorical determination. Minors, for

example, are incompetent in law, whereas adults can generally be declared legally

incompetent only on the basis of some factual determination. The issue of legal

capacity is more complex for adult patients, for whom an individual determination

normally must be made. If a person is incompetent, the physician is usually

required, absent an emergency, to secure some form of third-party consent from

a guardian or other legally empowered representative. Placement of the label

“incompetent” on a patient or subject automatically introduces the possibility of

coercive treatment and the presumption that there is no need to obtain consent.

“Competence” commonly functions to denote persons whose consents, refusals,

and statements of preference will be accepted as binding, while “incompetence”

denotes those who are to be placed under the guidance and control of another.

JUSTIFICATIONS FOR NOT OBTAINING CONSENT

Several standard exceptions to requirements of informed consent exist in law and

ethics. All courts passing judgment on the issue have ruled that a patient’s right to

self-determination is not absolute. Five exceptions to the informed consent

requirement are generally recognized: the public health emergency, the medical

emergency, the incompetent patient, the therapeutic privilege, and the patient

waiver (Meisel, 1979). All but the therapeutic privilege have been widely accepted

in law and ethics. However, the therapeutic privilege has elicited a particularly

furious exchange over whether autonomy rights can be validly set aside.

The therapeutic privilege allows a physician to withhold information based on a

sound medical judgment that to divulge the information would be potentially

harmful to a depressed, emotionally drained, or unstable patient. Several harmful

outcomes have been cited, including endangering life, causing irrational decisions,

and producing anxiety or stress (Canterbury v. Spence, 1972, 785–789; van Oosten,

1991, 31–41). In clinical settings, this privilege has long been used to justify not

obtaining consent. If framed broadly, the therapeutic privilege can permit physi-

cians to withhold information if disclosure would cause any countertherapeutic

deterioration, however slight, in the physical, psychological, or emotional condition

of the patient. If framed narrowly, it can permit the physician to withhold


information if and only if the patient’s knowledge of the information would have

serious health-related consequences—for example, by jeopardizing the success of

the treatment or harming the patient psychologically by impairing decisionmaking.

The narrowest formulation is that the therapeutic privilege can be validly

invoked only if the physician has good reason to believe that disclosure would

render the patient incompetent to consent to or refuse the treatment—that is,

would render the decision nonautonomous. To invoke the therapeutic privilege

under such circumstances does not conflict with respect for autonomy, because

an autonomous decision could not be made. However, broader formulations of

the privilege that require only “medical contraindication” do operate at the

expense of autonomy. These formulations may unjustifiably endanger autono-

mous choice, as when use of the privilege is based on the belief that an autono-

mous patient would refuse an indicated therapy for medically inappropriate

reasons.

Unless the therapeutic privilege is tightly and operationally formulated, the

medical profession can use it to deprive the unreasonable but competent patient

of the right to make decisions, especially if the physician sees his or her commit-

ment to the patient’s best interest as the overriding consideration. Loose stan-

dards can permit physicians to climb to safety over a straw bridge of speculation

about the psychological consequences of information. In short, there is a signifi-

cant potential for abuse of the privilege because of its inconsistency with the

patient’s rights to know and to decline treatment (Faden & Beauchamp, 1986, 38).

In 1986, U.S. Supreme Court Justice Byron White vigorously attacked the idea

that concerns about increasing a person’s anxiety about a procedure provide

grounds for an exception to rules of informed consent: “It is the very nature of

informed consent provisions that they may produce some anxiety in the patient

and influence her in her choice. This is in fact their reason for existence, and . . . it

is an entirely salutary reason” (Thornburgh v. American College of Obstetricians,

1986, 2199–2200). White is suggesting that the legal status of the doctrine of

therapeutic privilege is no longer as secure as it once was.

In addition to the five standard exceptions to informed consent, several cir-

cumstances are encountered in contemporary medicine and research that suggest

a need to relax requirements of informed consent. For example, in observational

studies that examine behavior without the subject’s knowledge and in secondary-

data analysis, we often need only to avoid invasions of privacy and the presenta-

tion of significant risk to subjects, without procuring consents. Some vital

research in epidemiology could not be conducted if consent were needed to

obtain access to records. Use of records without consent is not necessarily an

ethical violation. Research may be the first stage of an investigation to determine

whether one needs to trace and contact particular individuals and obtain their

permission for further participation in a study. In other cases, third-party consent


is sometimes acceptable when access to a subject is impractical or the subject is

incompetent. In some cases of low-risk research, subjects of research need not be

contacted at all, and occasionally disclosures and warnings may be substituted for

obtaining explicit informed consents. Thus, disclosures and warnings may some-

times justifiably be substituted for informed consents.

Finally, it does not follow from the great social importance of the rules and

practices of informed consent that institutional policies of informed consent must

rank the protection of decision making above all other values. The preservation of

autonomous choice is the first, but not the only, institutional commitment. For

example, a patient’s need for education and counseling must be balanced against

the interests of other patients and of society in maintaining a productive and

efficient health care system. Accordingly, institutional policies must consider

what is fair and reasonable to require of health care professionals and researchers

and what the effect would be of alternative consent requirements on efficiency and

effectiveness in the delivery of health care and the advancement of science.

CONCLUSION

Jay Katz has argued that the history of the physician–patient relationship from

ancient times to the present reveals how inattentive physicians have been to their

patients’ rights and needs. Katz is equally unrelenting in his criticisms of court

decisions and legal scholarship. He regards the declarations of courts as filled with

overly optimistic rhetoric. The problem, in his view, is that the law has little to do

with fostering real communication in the clinic and tends to line up with the

professional judgments of physicians in the crucial test cases (Katz, 1984, 2–4, 28,

49–50, 79; 1987).

Katz is correct in judging that informed consent has always been an alien notion

in the history of medicine and medical ethics and that informed consent require-

ments have still not deeply modified forms of communication between physicians

and their patients. At the same time, the scene in medicine throughout the world

is undergoing what may prove to be an extensive transformation through the

implementation of the idea of informed consent. Patients are giving more

“informed consents” and more attention is being paid in institutions to the

quality of those consents. It is indisputable that research ethics and policies

have been dramatically affected by requirements of informed consent.

Before we condemn the defects in the writings and practices of the past, we

should remember that the history of informed consent is still unfolding and that

our failures may be no less apparent to future generations than are the failures that

we find with the past.


References

Ackerman, Terrence F., and Carson M. Strong. In “Letters.” Journal of the American

Medical Association 249 (Feb. 18, 1983): 882–883.

American Medical Association, Proceedings of the National Medical Conventions. Held in

New York, May 1846, and in Philadelphia, May 1847. Adopted May 6, 1847, and

submitted for publication in Philadelphia, 1847.

Beauchamp, Tom L., and James F. Childress. Principles of Biomedical Ethics. 4th ed. New

York: Oxford University Press, 1994 (and later editions).

Berkey v. Anderson, 1 Cal. App. 3d 790, 82 Cal. Rptr. 67 (1969).

Bok, Sissela. “Informed Consent in Tests of Patient Reliability.” Journal of the American

Medical Association 267, no. 8 (Feb. 26, 1992): 1118–1119.

Bronstein, Alvin. “Remarks.” In Experiments and Research with Humans: Values in Conflict,

National Academy of Sciences, 130–131. Washington, DC: National Academy, 1975.

Buchanan, Allen E., and DanW. Brock.Deciding for Others: The Ethics of Surrogate Decision

Making. Cambridge: Cambridge University Press, 1989.

Canterbury v. Spence, 464 F.2d 772 (D.C. Cir. 1972).

Capron, Alexander M. In “Letters.” Journal of the American Medical Association 249 (Feb.

18, 1983): 882–883.

Cobbs v. Grant, 104 Cal. Rptr. 505, 502 P.2d 1 (1972).

Dubler, Nancy. “The Burdens of Research in Prisons.” IRB 4 (November 1982): 9–10.

Faden, Ruth R., and Tom L. Beauchamp. A History and Theory of Informed Consent. New

York: Oxford University Press, 1986.

Gisborne, Thomas. An Enquiry into the Duties of Men in the Higher and Middle Classes of

Society in Great Britain, resulting from their Respective Stations, Professions, and

Employments. London: B. & J. White, 1794.

Gregory, John. Lectures on the Duties and Qualifications of a Physician. London: W. Strahan

and T. Cadell, 1772.

Hagman, D. B. “The Medical Patient’s Right to Know: Report on a Medical-Legal-Ethical,

Empirical Study,” U.C.L.A. Law Review 17 (1970): 758–816.

Harris, Louis, et al. “Views of Informed Consent and Decisionmaking: Parallel Surveys of

Physicians and the Public.” In Making Health Care Decisions, Vol. 2, President’s

Commission for the Study of Ethical Problems in Medicine and Biomedical and

Behavioral Research (as below).

Hattori, Hiroyuki, et al. “The Patient’s Right to Information in Japan—Legal Rules and

Doctor’s Opinions.” Social Science and Medicine 32 (1991): 1007–1016.

Hershey, Nathan, and Bushkoff, Stanley H. Informed Consent Study. Pittsburgh, PA: Aspen

Systems Corporation, 1969.

Hippocrates. InHippocrates, 4 vols., translated byW.H.S. Jones, 1923-1931. Cambridge, MA:

Harvard University Press.

Hooker, Worthington. Physician and Patient; or a Practical View of the Mutual Duties,

Relations and Interests of the Medical Professions and the Community. New York: Baker

and Scribner, 1859.


Hyman v. Jewish Chronic Disease Hospital, 251N.Y. 2d 818 (1964), 206N.E. 2d 338 (1965).

Jones, James H. Bad Blood. 2nd ed. New York: Maxwell Macmillan, 1993.

Kai, Ichiro, et al. “Communication between Patients and Physicians about Terminal Care:

A Survey in Japan.” Social Science and Medicine 36 (1993): 1151–1159.

Katz, Jay. “Physician-Patient Encounters ‘On a Darkling Plain.’”Western New England Law

Review 9 (1987): 207–226.

Katz, Jay. The Silent World of Doctor and Patient. New York: The Free Press, 1984.

Levine, Robert J. “The Nature and Definition of Informed Consent in Various Research

Settings.” The Belmont Report. Appendix: Vol. I. Washington, DC: DHEW Publication

No. (OS) 78-0013, 1978, (3-1)–(3-91).

Levine, Robert J. “Research Involving Prisoners: Why Not?” IRB 4 (May 1982): 6.

Lipsett, Mortimer B. “On the Nature and Ethics of Phase I Clinical Trials of Cancer

Chemotherapies.” Journal of the AmericanMedical Association 248 (Aug. 27, 1982): 941–942.

Meisel, Alan. “The ‘Exceptions’ to the InformedConsent Doctrine: Striking a Balance Between

CompetingValues inMedicalDecisionmaking.”Wisconsin LawReview 1979 (1979): 413–488.

Meisel, Alan. “The Expansion of Liability for Medical Accidents: From Negligence to Strict

Liability by Way of Informed Consent.” Nebraska Law Review 56 (1977): 51–152.

Meisel, Alan, and Loren H. Roth. “Toward an Informed Discussion on Informed Consent:

A Review and Critique of the Empirical Studies.” Arizona Law Review 25 (1983): 265–346.

Meisel, Alan, and Loren Roth. “What We Do and Do Not Know About Informed Consent.”

Journal of the American Medical Association 246 (November 1981): 2473–2477.

Mizushima, Yutaka, et al. “A Survey Regarding the Disclosure of the Diagnosis of Cancer in

Toyama Prefecture, Japan.” Japanese Journal of Medicine 29 (1990): 146–155, esp. 146.

National Commission for the Protection of Human Subjects of Biomedical and

Behavioral Research. The Belmont Report. Washington, DC: DHEW Publication

No. (OS) 78-0012, 1978.

National Commission for the Protection of Human Subjects. Research Involving Prisoners.

Washington, DC: DHEW Publication No. (OS) 76-131, 1976.

Percival, Thomas. Medical Ethics; or a Code of Institutes and Precepts, Adapted to the

Professional Conduct of Physicians and Surgeons. Manchester, England: S. Russell, 1803.

The most available edition is Percival’s Medical Ethics, edited by Chauncey D. Leake.

Huntington, NY: Robert E. Krieger Publishing Company, 1975.

Pernick, Martin S. “The Patient’s Role in Medical Decisionmaking: A Social History of

Informed Consent in Medical Therapy.” In Making Health Care Decisions, Vol. 3,

President’s Commission for the Study of Ethical Problems in Medicine and Biomedical

and Behavioral Research. Washington, DC: U.S. Government Printing Office, 1982.

Planned Parenthood of Central Missouri v. Danforth, 428 U.S. 52 (1976).

President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and

Behavioral Research. Deciding to Forego Life-Sustaining Treatment. Washington, DC: U.S.

Government Printing Office, March 1983.

President’s Commission for the Study of Ethical Problems inMedicine and Biomedical and

Behavioral Research. Making Health Care Decisions, Vols. 1–3. Washington, DC: U.S.

Government Printing Office, 1982.


Proceedings of the Symposium on Ethical Issues in Human Experimentation: The Case of

Willowbrook State Hospital Research. Urban Health Affairs Program. New York:

University Medical Center, 1972.

Quaid, Kimberly A., et al. “Informed Consent for a Prescription Drug: Impact of Disclosed

Information on Patient Understanding and Medical Outcomes.” Patient Education and

Counselling 15 (1990): 249–259.

Rush, Benjamin. Medical Inquiries and Observations, Vol. 2, 2nd ed. Philadelphia: Thomas

Dobson, 1794. Published as a single essay entitled An Oration . . . An Enquiry into the

Influence of Physical Causes upon the Moral Faculty. Philadelphia: Charles Cist, 1786.

Salgo v. Leland Stanford Jr. University Board of Trustees, 317 P.2d 170 (1957).

Scherer, David G., and N. D. Reppucci. “Adolescents’ Capacities to Provide Voluntary

Informed Consent.” Law and Human Behavior 12 (1988): 123–141.

Schloendorff v. Society of New York Hospitals, 211N.Y. 125, 105N.E. 92 (1914).

Siminoff, L. A., and J. H. Fetting. “Factors Affecting Treatment Decisions for a Life-

Threatening Illness: The Case of Medical Treatment for Breast Cancer.” Social Science

and Medicine 32 (1991): 813–818.

Siminoff, L. A., J. H. Fetting, and M. D. Abeloff. “Doctor-Patient Communication about

Breast Cancer Adjuvant Therapy.” Journal of Clinical Oncology 7 (1989): 1192–1200.

Takahashi, Yoshimoto. “Informing a Patient of Malignant Illness: Commentary from a

Cross-Cultural Viewpoint.” Death Studies 14 (1990): 83–91.

Thornburgh v. American College of Obstetricians, 106 S.Ct. 2169 (1986) (White, J.,

dissenting).

United States v. Karl Brandt, Trials of War Criminals Before the Nuremberg Military

Tribunals under Control Council Law No. 10, Vols. 1 and 2, “The Medical Case,”

Military Tribunal I, 1947. Washington, DC: U.S. Government Printing Office, 1948–1949.

van Oosten, F. F. W. “The So-Called ‘Therapeutic Privilege’ or ‘Contra-Indication’: Its

Nature and Role in Non-Disclosure Cases.” Medicine and Law 10 (1991): 31–41.

Wilkinson v. Vesey, 295 A.2d 676 (R.I. 1972).


5

WHO DESERVES AUTONOMY AND WHOSE

AUTONOMY DESERVES RESPECT?

“Autonomy,” “respect for autonomy,” and “rights of autonomy” are strikingly

different notions. “Respect for autonomy” and “rights of autonomy” are moral

notions, but “autonomy” and “autonomous person” are not obviously moral

notions. Indeed, they seem more metaphysical than moral. However, this dis-

tinction between the metaphysical and the moral has fostered precarious claims

such as these: (1) analysis of autonomy is a conceptual, metaphysical project, not a

moral one; (2) a theory of autonomy should not be built on moral notions, but on

a theory of mind, self, or person; and (3) the concept of autonomy is intimately

connected to the concept of person, which anchors the concept of moral status.

I will be assessing these claims with the objective of determining who qualifies

as autonomous and what sort of autonomy deserves our respect. I will argue that

moral notions—in particular, respect for autonomy—should affect how we con-

struct theories of autonomous action and the autonomous person. However,

theories of autonomy should only be constrained by the principle of respect for

autonomy, not wholly determined by it.

CONCEPTS AND THEORIES OF AUTONOMY

Autonomy is generally understood as personal self-governance: personal rule of

the self free of controlling interferences by others and free of personal limitations

that prevent choice. Two basic conditions of autonomy, therefore, are (1) liberty

(independence from controlling influences) and (2) agency (capacity for inten-

tional action). However, disagreement exists over how to analyze these conditions

and over whether additional conditions are needed. Each notion is indeterminate

until further specified, and each can be used only as a rough guide for philoso-

phers in the construction of a theory of autonomy.1

Some available theories of autonomy feature traits of the autonomous person,

whereas others focus on autonomous action. Theories of the autonomous person

are theories of a kind of agent. For example, the autonomous person is por-

trayed in some theories as consistent, independent, in command, resistant to

control by authorities, and the source of his or her basic values and beliefs.

These theories are structured in terms of virtues and persistent ideals. My

analysis of autonomy is not focused on such traits of the person, but on actions.

My interest is on choice rather than general capacities for governance. Until the

final section in this essay, I will not be concerned with conditions of person-

hood. I assume that autonomous persons sometimes fail to act autonomously

because of temporary constraints caused by illness or depression, circumstantial

ignorance, coercion, or other conditions that restrict options. An autonomous

person who signs a contract or consent agreement without understanding the

document and without intending to agree to its conditions is qualified to act

autonomously, but fails to do so. Similarly, a man who is threatened with death

by a thief and who hands over his wallet because he does not wish to suffer the

threatened consequences does not act autonomously, even if he is an autono-

mous person.

Conversely, some persons who are generally incapable of autonomous decision

making make some autonomous choices. For example, some patients in mental

institutions who are unable to care for themselves and are legally incompetent

make autonomous choices such as ringing for a nurse, stating preferences for

meals, and making telephone calls to acquaintances. These persons act autono-

mously even if they fail critical conditions of the autonomous person.

THE ROLE OF THE PRINCIPLE OF RESPECT FOR AUTONOMY

To maintain coherence with fundamental principles of morality, a theory of

autonomy should be kept consistent with the substantive assumptions about

autonomy implicit in the principle of respect for autonomy. To respect an auton-

omous agent is to recognize with due appreciation that person’s capacities and

perspective, including the right to control his or her affairs, tomake certain choices,

and to take certain actions based on personal values and beliefs. Autonomous

agents are entitled to determine their own destiny, and respect requires noninter-

ference with their actions. Respect involves acknowledging decision-making rights


and enabling persons to act, whereas disrespect involves attitudes and actions that

ignore, insult, or demean others’ rights of autonomy. This thesis does not entail

that, from the moral point of view, we are to respect only the morally good

intentions and actions of agents. Many acts of individual autonomy are morally

neutral, yet are owed respect.

Making Theory Conform to Moral Principle

Some theories of autonomy do not presume that a principle of respect for

autonomy provides a substantive basis for the theory. They do not mention the

principle or attempt to conform the theory to its assumptions. This matter is of

the first importance because a theory that distinguishes nonautonomous acts

from autonomous ones teaches us what it is that we are to respect, and opens

up the possibility of disrespecting certain “choices” that are of penetrating

importance to the agent, on grounds that these choices are nonautonomous.

Any theory that classifies acts as not autonomous that are of the greatest

importance to us in the basic governance of our affairs is morally dangerous

and conceptually dubious. If, for example, a theory declares nonautonomous

the acts of average persons in opening a bank account, writing a will, selling a

house, or refusing an offered surgical procedure, the theory is unacceptable. To

declare these choices nonautonomous is to imply that another person may

legitimately serve as guardian and decision maker. On this basis a will could be

invalidated or a surgical procedure authorized against the person’s wish. As

Onora O’Neill has cautioned in her work, standards of autonomy so high that

we cannot live up to them will have the devastating effect of inappropriately

classifying “a far larger proportion of the patient population as lacking

competence.”2

An instructive example of the moral perils that a theory of autonomy can pose

is found in the work of Julian Savulescu on decisions to limit choices for or

against life-sustaining treatments. In developing his account, Savulescu realizes

that he must set out the conditions of autonomy in just the right way in order to

get morally justified outcomes. If he fails in the theory of autonomy, the choices

of patients will be inappropriately limited or reversed. Savulescu sharply distin-

guishes autonomous and nonautonomous acts using a distinction betweenmere

desires and rational desires. Autonomous actions are only those performed from

rational desires. Savulescu argues that health care professionals and guardians

are only required to respect the actions of a patient that are done from rational

desire; an expressed desire is not sufficient. He argues that many choices (e.g., a

Jehovah’s Witness’s decision to refuse a blood transfusion) can and should be

WH O D E S E R V E S A U T O N OM Y ? 81

judged not rational, and therefore not autonomous, and therefore lacking in

moral weight.3

The theory of autonomy that I present in this essay presumes the cardinal moral

importance of protecting everyday choices by consulting the principle of respect

for autonomy. I am assuming that everyday choices of ordinary persons are

paradigm cases of autonomous choices. A critical test of the adequacy of any

theory of autonomy, then, is whether it coheres with the moral requirement that

we respect everyday choices such as opening bank accounts, purchasing goods in

stores, and authorizing an automobile to be repaired.

I am not asserting either that ordinary persons never fail to choose autono-

mously or that we are morally required to respect all autonomous choices by not

interfering with them. Certain behaviors “performed” or “willed” by individuals

are clearly not autonomous (e.g., giving up one’s wallet when coerced and

attempting to fly off a balcony when in a drug-induced state); and we are just as

certainly not required to respect all choices of thieves, religious zealots who spew

hatred, persons who act with conflicts of interest, and the like.

A Typical Example

As an example of the choices of ordinary persons having status as autonomous,

consider the refusal of a blood transfusion by a Jehovah’s Witness who has never

reflectively questioned whether he should be a member of his faith or asked

whether he wants to be the kind of person who refuses blood transfusions.

Throughout his life he has been a firmly committed Jehovah’s Witness, but now

his religious commitments conflict with the healing commitments of health care

professionals who are urging a transfusion. His life is on the line, and he adheres to

the doctrines of his faith.

One could challenge the proposition that this Jehovah’s Witness is acting

autonomously on grounds that his beliefs are unreflective assumptions instilled

by authoritarian dogma or what Savulescu calls irrational desires, but this chal-

lenge seems to me conceptually and empirically doubtful and, as a matter of

theory and policy, fraught with danger. That we adopt beliefs and principles

deriving from forms of institutional authority does not prevent them from

being our beliefs and principles. Individuals autonomously accept moral notions

that derive from many forms of cultural tradition and institutional authority. If

the Witness’s decision can be legitimately invalidated on grounds of acting

nonautonomously, so may a great many institutionally guided choices. I hypothe-

size that a theory of autonomy that conflicts with this assumption is an unac-

ceptable and morally problematic theory.


A CONCISE THEORY OF THE CONDITIONS OF AUTONOMOUS ACTION

I will be assuming in this article an account of autonomous choice that I have

elsewhere set out in terms that I believe to be compatible with the constraints on

theory that I just outlined.4 That is, this account of autonomy is designed to be

coherent with the assumptions that we must make when we insist that the choices

of ordinary persons be respected. In this account, I analyze autonomous action in

terms of normal choosers who act (1) intentionally, (2) with understanding, and

(3) without controlling influences. What follows is a brief statement of these

conditions, omitting their subtleties.

The Condition of Intentionality

Intentional actions require plans in the form of representations of the series of

events proposed for the execution of the action. For an act to be intentional, it

must correspond to the actor’s conception of the act in question, although a

planned outcome might not materialize as projected. Unintended acts, such as a

pediatrician’s dropping of a newborn infant during delivery, are nonautonomous,

but the agent may still be held responsible for what occurs. Imagine that Professor

X intends to read a paper that he has written on a toxic chemical and intends

thereby to win first prize at a convention. His paper is on toxic chemical A.

Professor X’s plan entails executing a sequence of actions, chief of which is a

particular set of arguments about A that he believes will be well received by this

audience. However, in preparing to travel to the convention, Professor X acci-

dentally packs the wrong paper, which is on the subject of toxic chemical B. The

professor is forgetful and does not realize when he reads his paper at the conven-

tion that it is the wrong paper. Despite this lapse, he wins first prize for the best

paper. Professor X has read a paper on a toxic chemical and won first prize, which

he intended, yet he did not do so intentionally in the way the action was

performed. His reception by the audience and his winning the prize were, as it

turns out, not brought about according to his plan of action.

In a theory of autonomy (and of informed consent), we need to decide whether

special kinds of wants or desires are necessary conditions of intentional action. It

might be thought that foreseen acts that the actor does not want and does not desire

to perform are not intentional. Alvin Goldman uses the following example in an

attempt to prove that agents do not intend merely foreseen effects.5 Imagine that

Mr. G takes a driver’s test to prove competence. He comes to an intersection that

requires a right turn and extends his arm to signal for a turn, although he knows it is

raining and that he will get his hand wet. According to Goldman, Mr. G’s signaling

for a turn is an intentional act. By contrast, his wet hand is an unintended effect, or

W H O D E S E R V E S A U T O N OM Y ? 83

“incidental by-product,” of his hand signaling. However, as I see it, getting the hand

wet is part of an intentional action. It is willed in accordance with a plan and is

neither accidental, inadvertent, nor habitual. There is nothing about intentional

acts, as opposed to accidental occurrences, that rules out aversions. One’s motiva-

tion for getting one’s hand wet may reflect conflictingwants and desires, and clearly

the driver does not want a wet hand, but this fact does not render the act of getting

the hand wet less than intentional or autonomous.

Some intentional acts are wanted or desired for their own sake, and not for the

sake of something else. For example, someone who loves to swim may want to

swim for the sake of swimming, as may not be the case for another person who

swims for the sake of a tanned or fit body. The desire to perform an act for its own

sake is intrinsic wanting. Wanting for the sake of something else is instrumental

wanting. When a surgeon cuts into the body of the patient, the patient has

consented to the cutting and the cutting is part of the plan, but it is for the sake

of better function or health. In such cases, the actor believes the action is the

means to a goal wanted in the primary sense. Goldman would only consider an act

intentional if at least one of these two forms of wanting is involved.

However, other intentional acts are not wanted in either of these two senses. An

actor may view these acts as, ceteris paribus, undesirable or unwanted. They are

performed only because they are entailed in the doing of other acts that are

wanted. Such acts are foreseen and wanted acts in the sense that the actor wants

to perform them in the circumstances more than he or she wants not to perform

them. It is suitable in such cases to discard the language of “wanting” and to say

that foreseen but not desired effects are “tolerated.”6 These effects are not so

undesirable that the actor would choose not to perform the act in the first place.

The actor includes them as a part of his or her plan of intentional action.

Accordingly, I use a model of intentionality based primarily on what is willed

rather than what is wanted. Intentional actions include any action and any effect

specifically willed in accordance with a plan, including merely tolerated effects.7 In

this conception, a physician can desire not to do what he intends to do, in the

same way that one can be willing to do something but reluctant to do it, or even

detest doing it. Consider Mr. X, who has become convinced that there is no

meaningful way that he can both undergo facial surgery and avoid the scarring

involved. After considering the alternative of refusing surgery, Mr. X intentionally

consents to surgery, and in so doing he intentionally consents to being scarred by

surgery. Although his consenting to a facial scar is a toleration of and not a desire

for the scarring, the intentional act of consenting to the scarring is no less Mr. X’s

own act than is his consenting to surgery. Mr. X, then, intentionally consents to

being scarred by surgery.

Under this account, the distinction between what agents do intentionally and

what they merely foresee in a planned action is not viable.8 For example, if a man


enters a room and flips a switch that he knows turns on both a light and a fan, but

desires only to activate the light, he cannot say that he activates the fan unin-

tentionally. Even if the fan made an obnoxious whirring sound that he does not

want, it would be incorrect to state that he unintentionally brought about the

obnoxious sound by flipping the switch. Here and in the other examples men-

tioned previously, one might introduce a distinction between what an agent

intends and an intentional action. Perhaps the agent does not intend the effect

of a screeching fan and does not intend it as the goal of his action even though it is

an intentional action. Likewise, one might be said to consent intentionally to

being scarred in surgery while not intending to be scarred. In short, what it means

for an act to be intentional need not be equated with what the agent of the action

intends. However, I will not further pursue the merits of this distinction between

what is intentional and what is intended, because I am concerned only with

analysis of what it means for an action to be intentional, and especially the

relationship between intentional action and autonomous choice.

The Condition of Understanding

An action is not autonomous if the actor has no appropriate understanding of it.

Here we need a way of analyzing the question, “Do you understand what you are

doing?” Starting with the extreme of full or complete understanding, a person

understands an action if the person correctly apprehends all of the propositions

that correctly describe the nature of the action and the foreseeable consequences

or possible outcomes that might follow as a result of performing or not per-

forming the action. This full understanding does not amount to omniscience,

because the criterion demands only foreseeability. Less complete understanding

occurs by degrees, with no understanding at the end of the continuum.

There are several reasons why limited understanding can occur in a process of

deliberative choice and consent. Some patients and subjects are calm, attentive,

and eager for dialogue, whereas others are nervous or distracted in ways that

impair or block understanding. Conditions that limit their understanding include

illness, irrationality, and immaturity. Deficiencies in the communication process

also often hamper understanding. A breakdown in a person’s ability to accept

information as true or untainted, even if he or she adequately comprehends the

information, can compromise decision making. For example, a seriously ill

patient who has been properly informed about the nature of the illness and has

been asked to make a treatment decision might refuse the proposed treatment

while under the false belief that he or she is not ill. Even if the physician recognizes

the patient’s false belief and adduces conclusive evidence to prove to the patient

that the belief is mistaken, and the patient comprehends the information


provided, the patient still may go on believing that what has been (truthfully)

reported is false.

The so-called “therapeutic misconception” occurs when subjects in research

fail to distinguish between clinical care and research and fail to understand the

purpose and aim of research, misconceiving their participation as therapeutic in

nature.9 Such a therapeutic misconception may invalidate the subject’s consent to

research. Some participants understand that they are involved in research, and not

clinical care, but overestimate the therapeutic possibilities and probabilities—that

is, the odds that a participant will benefit.10 This overestimation may render a

choice insufficiently autonomous and invalidate a consent.

Autonomous actions require only a basic understanding of the action, not a full

understanding. An account of understanding that required an extremely high

level of understanding would be oppressive if made a condition of autonomy.

Many years ago when I decided to become a philosopher, I did not well under-

stand the profession or its demands. However, I did know something about the

writings of philosophers, I did know a few graduate students in philosophy, and I

did have a reasonably good idea of my abilities as a student of philosophy. To say

that my decision to be a philosopher was nonautonomous because I lacked

relevant information about the profession would be inaccurate, even if such

information was relevant to the choice made.

The Condition of Noncontrol, or Voluntariness

The third of the three conditions constituting autonomy is that a person, like an

autonomous political state, must be free of controls exerted either by external

sources or by internal states that rob the person of self-directedness. Influence

and resistance to influence are basic concepts for this analysis. Not all influences

are controlling. Many influences are resistible, and some are even trivial in their

impact on autonomy. The category of influence is broad enough to include acts

of love, threats, education, lies, manipulative suggestions, and emotional

appeals, all of which can vary dramatically in their impact on persons. My

analysis will focus on three categories of influence: persuasion, coercion, and

manipulation.11

Persuasion is here understood as rational persuasion. A person comes to believe

something through the merit of reasons another person advances. This is the

paradigm of an influence that is not controlling and also warranted. If a physician

presents reasons that persuade the patient to undergo a procedure when the

patient is reluctant to do so, then the physician’s actions influence, but do not

control. In health care settings, there is often a problem of distinguishing emo-

tional responses from cognitive responses and of determining whether the


cognitive or the emotional is the primary factor in a decision. Some informational

approaches that rationally persuade one patient overwhelm another whose fear or

panic short-circuits reason.

Coercion occurs if and only if one person intentionally uses a credible and severe

threat of harm or force to control another.12 If a physician orders a reluctant

patient to undergo a diagnostic examination and coerces the patient into com-

pliance through a threat of abandonment, the physician’s influence controls the

patient. The threat of force used by some police, courts, and hospitals in acts of

involuntary commitment for psychiatric treatment is coercive. Some threats will

coerce virtually all persons (e.g., a credible threat to imprison a person), whereas

others might succeed in coercing only a few persons (e.g., a parent’s threat to cut

off funding for a son in college unless the son consents to the donation of one of

his kidneys to his sister).

Compliance merely because a person feels threatened, when no threat has been

issued, is not coercion. Coercion occurs only if a credible and intended threat

disrupts and reorders a person’s self-directed course of action. Under these coercive

conditions, even intentional and well-informed actions can be nonvoluntary.

Handing over one’s wallet to a thief with a gun is both intentional and well

informed, though nonvoluntary. Being coerced does not entail that the person

coerced lacks voluntary decision-making capacity in complying with a threat. It also

does not entail that the person did not choose to perform the action or that the

coercion involved always invalidates consent. The point is only that the action was

not voluntary.

Manipulation is a term for several forms of influence that are neither persua-

sive nor coercive. It involves getting people to do what the manipulator wants

through a nonpersuasive means that alters a person’s understanding of a situa-

tion and motivates the person to do what the agent of influence intends.

In health care and research, the most likely forms of manipulation are informa-

tional manipulation that nonpersuasively alters a person’s understanding and

offers of rewards or potential benefits. Critics of various recruiting practices in

biomedical research have suggested that there is often informational manipula-

tion through withholding critical information about risks together with a mis-

leading exaggeration of benefits. Other critics have said that offers of

compensation and health care are manipulative when made excessively

attractive.

Nevertheless, it is easy to inflate the threat of control by manipulation beyond

its significance in health care. We typically make decisions in a context of

competing influences such as familial constraints, legal obligations, offers of

rewards, and institutional pressures. These influences usually do not control

decisions to a morally questionable degree. In biomedical ethics we need only

establish general criteria for the point at which influence threatens autonomous


choice, while recognizing that in many cases no sharp boundary separates con-

trolling and noncontrolling influences.

I have concentrated in this section on external controlling influences—usually

attempts by one person to influence another. No less important are internal

influences on the person, such as those caused by mental illness. These conditions

can also be voluntariness depriving. However, I will not here pursue problems of

lack of internal control. I simply stipulate that an adequate theory of voluntariness

must take account of both internal and external controlling influences.

Degrees of Autonomy and Substantial Autonomy

The first of the aforementioned three conditions of autonomy, intentionality, is

not a matter of degree: Acts are either intentional or nonintentional. However,

acts can satisfy both the conditions of understanding and absence of controlling

influences to a greater or lesser extent. For example, threats can be more or less

severe, and understanding can be more or less complete. Actions are autono-

mous by degrees, as a function of satisfying these conditions to different degrees.

For both conditions, a continuum runs from fully present to wholly absent. For

example, children exhibit different degrees of understanding at various ages, as

well as different capacities of independence and resistance to influence attempts.

This claim that actions are autonomous by degrees is an inescapable conse-

quence of a commitment to the view that two of the necessary conditions of

autonomy are satisfied by degrees.

For an action to be classified as either autonomous or nonautonomous, cut-

off points on these continua are required. To fix these points, only a substantial

satisfaction of the conditions of autonomy is needed, not a full or categorical

satisfaction of the conditions. The line between what is substantial and what is

insubstantial may seem arbitrary, but thresholds marking substantially auton-

omous decisions can be carefully fixed in light of specific objectives of decision

making, such as deciding about surgery, buying a house, choosing a university

to attend, making a contribution to charity, driving a car, and hiring a new

employee.

Problems of Adequacy and Completeness

One can, and should, raise questions about whether the theory I have outlined

contains an adequate set of conditions for a theory of autonomy. This theory

looks to be what, in other philosophical literature, might be called a theory of free

will or free agency. Free agency has often been analyzed in terms of (1) a


condition of persons through which intentional actions are willed (a condition

absent in contrastive cases such as dreaming and unintentional word-slips) and

(2) the absence of internal and external controls (compulsion, constraint, etc.)

that determine the choice of actions. This theory of free choice is similar to the one

I am proposing for the analysis of autonomy.

However, if my theory reduces to a theory of free agency, it may have missed its

target of autonomy. To test this hypothesis, I now look at a type of theory that

proposes one or more conditions different from those that I have proposed.

SPLIT-LEVEL THEORIES OF AUTONOMY

Several philosophersmaintain that autonomy consists of the capacity to control and

identify with one’s first-order desires or preferences by means of higher-level

(second-order) desires or preferences through processes of deliberation, reflection,

or volition. Harry Frankfurt’s theory of the freedom of persons and Gerald

Dworkin’s theory of autonomy are widely discussed examples. I will closely follow

the language ofDworkin’s theory, since Frankfurt’s theory of persons is not explicitly

presented as a theory of autonomy.13However, I give neither theory a priority status.

An autonomous person, in this theory, is one who has the capacity to accept,

identify with, or repudiate a lower-order desire or preference, showing the

capacity to change (or maintain) one’s preference structure or one’s configura-

tion of the will. All and only autonomous persons possess such distanced self-

reflection, in which second-order mental states have first-order mental states as

their intentional objects and considered preferences are formed about first-

order preferences and beliefs. For example, a long-distance runner may have a

first-order desire to run several hours a day, but also may have a higher-order

desire to decrease the time to one hour. If he wants at any given moment to run

several hours, then he wants at that moment what he does not truly want. Action

from a first-order desire that is not endorsed by a second-order volition is not

autonomous and is typical of animal behavior.

Frankfurt argues that it is essential to being a person that the second-order desires

or volitions be such that the individual “wants a certain desire to be his will.”14

These second-order desires or volitions he calls “second-order volitions.” Since they

are essential to being a person, any individual lacking these volitions is not a person.

Dworkin offers a “content-free” definition of autonomy as a “second-order capacity

of persons to reflect critically upon their first-order preferences, desires, wishes, and

so forth and the capacity to accept or attempt to change these in the light of higher-

order preferences and values.”15 The language of capacity strongly suggests that this

theory is one of autonomous persons, not a theory of autonomous actions.


Problems with the Theory

Several problems haunt this theory. First, there is nothing to prevent a reflective

acceptance, preference, concern, or volition at the second level from being caused

by and assured by the strength of a first-order desire. The individual’s acceptance

of or identification with the first-order desire would then be nomore than a causal

result of an already formed structure of preferences. Frankfurt writes that

“whether a person identifies himself with [his] passions, or whether they occur

as alien forces that remain outside the boundaries of his volitional identity,

depends upon what he himself wants his will to be.”16 The problem is that the

identification with one’s passions may be governed by the strength of the first-

order passion, not by an independent identification. If a person’s identification

(from what “he himself wants his will to be”) at any point is itself the result of a

process of thoroughgoing conditioning or lower-level passion, then the identifi-

cation is never sufficiently independent to qualify as autonomous.

For example, the alcoholic with a passion for red wine who identifies with

drinking seems nonautonomous if his second-level volition or desire to drink red

wine is causally determined by a first-level desire. Suppose the alcoholic forms, as

a result of the strength of first-order desire, a second-order volition to satisfy his

strongest first-order desire, whatever it is. This behavior seems nonautonomous,

but looks as if it would satisfy Frankfurt’s conditions. Moreover, an alcoholic can

reflect at ever-higher levels on lower-level desires without achieving autonomy if

identification at all levels is causally determined by initial desires.

To make this split-level theory plausible as an account of autonomy, a supple-

mentary theory would have to be added that distinguished influences or desires

that rob an individual of autonomy from influences or desires consistent with

autonomy.17 Frankfurt seems to address these problems with the thesis that “truly

autonomous choices” require “being satisfied with a certain desire” and having

preexisting “stable volitional tendencies.”18However, this analysis does not rescue

the split-level theory. Anomalous actions can derive from choices that are out of

character as a result of surrounding events that are unprecedented in the actor’s

experience, such as serious disease. These acts can be well planned, intentional,

and free from the control of other persons. The actors may be unaware of the

motivational or conditioning history that underlies and prompts their actions and

may have made no reflective identification with the origins of their actions. This

fact does not supply a sufficient reason to declare them nonvoluntary or

nonautonomous.

Second, this theory risks running afoul of the criterion of coherence with the

principle of respect for autonomy mentioned earlier. If reflective identification

with one’s desires or second-order volitions is a necessary condition of autono-

mous action, then many ordinary actions that are almost universally considered


autonomous, such as cheating on one’s spouse (when one truly wishes not to

be such a person) or selecting tasty snack foods when grocery shopping (when one

has never reflected on one’s desires for snack foods), would be rendered non-

autonomous in this theory. Requiring reflective identification and stable voli-

tional patterns unduly narrows the scope of actions protected by a principle of

respect for autonomy.

Frankfurt’s theory runs this risk in its treatment of persons who have a

“wanton” lack of concern about “whether the desires that move him are desires

by which he wants to be moved to act.” Such an individual, says Frankfurt, is “no

different from an animal.”19 Indeed, “insofar as his desires are utterly unreflective,

he is to that extent not genuinely a person at all. He is merely a wanton.”20 This

theory needs more than a convincing account of second-order desires and voli-

tions; it needs a way to ensure that ordinary choices qualify as autonomous even

when persons have not reflected on their preferences at a higher level and even

when they are hesitant to identify with one type of desire rather than another.

Depending on how “reflection,” “volition,” and the like are spelled out in this

theory, few choosers and few choices might turn out to be autonomous because

few would fail to engage in higher-order reflection. Often the agents involved will

not have reflected on whether they wish to accept or identify with the motiva-

tional structures that underlie such actions. Actors will in some cases be unaware

of their motivational or conditioning histories and will have made no reflective

identifications. Actions such as standing up during a religious service, lying to

one’s physician about what one eats, and hiding one’s income from the Internal

Revenue Service might on this basis turn out to be nonautonomous. The moral

price paid in this theory is that individuals who have not reflected on their desires

and preferences at a higher level deserve no respect for actions that derive from

their most deep-seated commitments, desires, and preferences. There is a danger

that they will be classified as no different than animals.21

Conversely, if one relaxes the standards of higher-order reflection, then many

acts will become autonomous that these theorists wish to exclude from the realm

of autonomy. For example, some actions of nonhuman animals will be autono-

mous. I will return to this problem in the final section of this essay.

Converting the Theory to Nonrepudiated Acceptance

The defender of a split-level theory could shift ground and require as a condition

of autonomous action only nonrepudiation in the values underlying choice, not

reflective acceptance of them—a move occasionally suggested, obliquely, by

Frankfurt.22 This position is negative rather than positive: Values, motives, and

actions are autonomous if the agent does not reflectively repudiate or abjure


them, and not autonomous if they are repudiated. This set of repudiated actions

would presumably not be large, but would include important cases of weakness of

the will, such as acts of taking drugs and acts of infidelity in which the person

repudiates the driving desire or value while nonetheless acting on it: “I was

seduced in a weak moment.” This position does not seem to make the mistake

of rendering nonautonomous most of our ordinary actions that are intentional

and informed.

Intriguing illustrations of this thesis come from clinical examples of repudiated

phobic and compulsive behavior such as the repudiation of a compulsive hand

washer. I concede the attractiveness of the theory for these examples, but I push

beyond them to more commonplace examples of repudiated action to see if the

theory remains attractive. What should we say about a corporate executive who

sincerely repudiates her characteristic avarice and greed? She repeatedly, but

unsuccessfully, tries to become more spiritual and less material in her desires.

She goes on being an aggressive corporate executive of the sort that she wishes she

were not, but is. Are her actions nonautonomous because repudiated? Her

repudiation is an insufficient reason to withhold classification of her acts as

autonomous. The perpetual dieter who continuously repudiates eating carbohy-

drates but goes on eating carbohydrates anyway is a similar case. These agents may

not be acting autonomously, but, if not, nonautonomy seems to derive from some

form of involuntariness (e.g., uncontrollable desire), not from a failure to con-

form to a repudiation of desire.

There are related problems about allowing a person’s actions to qualify as

autonomous when values and motives are not repudiated. For example, in the

case of the compulsive hand washer, suppose that, instead of repudiating desires for

or the choice of hand washing, the handwasher has never reflectively considered his

or her desires or motives. It is implausible to describe compulsive actions of hand

washing as autonomous. Again, we have a problem of external noncontrol, not

internal nonrepudiation. Similar questions can be raised about Frankfurt’s

unwilling addict who identifies himself through a second-order volition as desiring

not to be an addict but goes on being one because he is in the hold of the addiction.

Frankfurt argues that this addict is a person, not a wanton, but that his actions of

drug intake are not of his own free will.23 I agree that he is a person who is not acting

freely, but is he acting autonomously? Frankfurt never answers this question, but

the analysis suggests (whether or not this conclusion is intended) that the addict

“acts” unfreely but autonomously.

It will not be easy to construct a split-level theory using second-order identi-

fication or nonrepudiation as a criterion. These conditions are not clearly needed

for a theory of autonomous action or the autonomous person. The condition of

noncontrol (as analyzed previously) is a more promising way to fill out a theory of

both autonomous choice and the autonomous person.


AUTONOMOUS PERSONS AND THE PROBLEM OF MORAL STATUS

Thus far I have concentrated on autonomous choice, although the most plausible

interpretation of split-level theories (and other theories of autonomy) is that they

are theories of autonomous persons. Setting aside how best to interpret particular

theories, I proceed in this section to autonomous persons in order to address

questions of moral status.

It is often assumed in philosophical literature on persons that they are essen-

tially autonomous; all persons are autonomous, and all autonomous individuals

are persons. However, not all theories make such an assumption. Some maintain

that fetuses, young children, advanced Alzheimer’s patients, and even the irrever-

sibly comatose are persons, but ones who lack autonomy. I will not pursue this

controversy. My concern is with theories that find autonomy or some feature of

autonomy (e.g., intention, rationality, or second-level identification) to be con-

stitutive of personhood. When I use the term person hereafter, it refers exclusively

to autonomous persons.

I said earlier that a theory with demanding conditions of autonomy (e.g.,

requiring full understanding, no external or internal constraint on the agent, or

second-level reflective identification) will reach the conclusion that many of

our presumed choices are not autonomous choices. This conclusion likewise

applies to persons; the more demanding the conditions in a theory of persons,

the fewer the number of individuals who satisfy the conditions, and therefore

the fewer who qualify for the moral status conferred by being persons. The

conditions in the theory can then be built into how we should interpret the

principle of respect for autonomy (or, perhaps, respect for persons). As the

quality or level of required mental skill is reduced in a theory, the number of

individuals who qualify for protection under the principle will increase. As the

quality or level of mental activity is increased (made more demanding), the

number of individuals who qualify for protection under the principle will

decrease.

If a theory demands a high threshold of mental capacity and a robust

personal history of reflective identification with values, then various indivi-

duals normally regarded as autonomous will be deemed nonautonomous, or

at least many of their preferences and choices will be rendered nonautonomous.

For example, many decisions by hospital patients about their care will be

classified as nonautonomous.24 Correlatively, if a theory demands only a low

threshold of mental skills (modest understanding, weak resistance to manip-

ulation, etc.), then many individuals who are normally regarded as nonautono-

mous will be deemed autonomous—for example, certain nonhuman animals.

This problem underlies and motivates my discussion of moral status in this

section.


Metaphysical and Moral Theories

Some theories of autonomous persons aremetaphysical, othersmoral.25 As I draw

the distinction, autonomous persons in a metaphysics of persons are identified by

a set of psychological (not moral) properties. Properties found in various theories

include intentionality, rationality, self-consciousness (of oneself as existing over

time), free will, language acquisition, higher-order volition, and possibly various

forms of emotion.26 The metaphysical goal is to identify a set of psychological

properties possessed by all and only autonomous persons. Morally autonomous

persons, by contrast, are capable of moral agency. In principle, an entity could

satisfy all the properties requisite for being a metaphysically autonomous person

and lack all the properties requisite for being a morally autonomous person.

Unfortunately,most theories of persons cannot easily be distinguished into one of

these two types. These theories are not attentive to the distinction between meta-

physical and moral persons. For three decades, and arguably for several centuries,

the dominant trend in the literature has been todelineate properties of individuals in

a metaphysical account from which conclusions can be drawn about their moral

status.Most philosophical accounts attempt to remain faithful to the commonsense

concept of person, which is, roughly speaking, identical to the concept of human

being. However, there is no warrant for the assumption that only properties

distinctive of the human species count toward personhood or autonomy, or that

species membership has anything to do withmoral status. Even if certain properties

that are strongly correlated with membership in the human species qualify humans

formoral statusmore readily than themembers of other species, these properties are

only contingently connected to being human. The properties could be possessed by

members of nonhuman species or by entities outside the sphere of natural species

such as the great apes and genetically manipulated species.27

What Have Metaphysical Theories to Do with Moral Theories?

Proponents of metaphysical theories often spread confusion by moving from a

metaphysical claim about persons to one about moral status or moral worth. This

move can be baffling, because metaphysical properties have no moral implica-

tions. A metaphysical-to-moral connection can be made only through a correla-

tive appeal to amoral principle such as a principle of respect for persons or respect

for autonomy. The principle must be defended independently of the metaphysical

theory and given some suitable content and relationship to the theory. If indivi-

dual X acts autonomously, rationally, self-consciously, and the like, how is moral

autonomy or any form of moral status established by this fact? No moral conclu-

sions follow merely from the presence of these properties.


Many philosophers, including Aristotle and Kant, hold that animals have

minds but lack critical human properties such as rationality, language use, or

dignity. Kant judged that human dignity—which he closely linked to moral

autonomy—places humans in a privileged position in the order of nature;

humans have properties that confer upon them a moral status not held by

nonhuman animals.28 Whatever the merits of Kant’s view in particular, the

belief persists generally in philosophy, religion, and popular culture that some

special property—perhaps autonomy or a property connected to it—confers a

unique moral status or standing on human persons. In philosophy, it is com-

monly asserted that nonhuman animals lack properties such as self-awareness, a

sense of continuity over time, the capacity to will, the capacity to love, and/or

autonomy, and therefore lack personhood (or its functional equivalent), and

therefore lack moral status.29

However, it is not demonstrated in these theories that nonhuman animals in

fact lack the relevant property. I have not seen a philosophical theory that argues

the point by reference to available empirical evidence. A typical statement by

philosophers is the following thesis of Frankfurt’s:

It is conceptually possible that members of novel or even of familiar non-

human species should be persons. . . . It seems to be peculiarly characteristic

of humans, however, that they are able to form what I shall call “second-

order desire”. . . . Many animals appear to have the capacity for what I shall

call “first-order desires”. . . .No animal other than man, however, appears to

have the capacity for reflective self-evaluation that is manifested in the

formation of second-order desires.30

Philosophical theories typically ignore striking evidence of types and degrees of

self-awareness of nonhuman animals, not to mention the pervasive presence of

intentionality in animals and the findings in comparative studies of the brain. In

some studies, language-trained apes appear to make self-references, and many

animals learn from the past and then use their knowledge to forge intentional

plans of action.31 These animals distinguish their bodies and interests from those

of the bodies and interests of others. In play and social life, they understand

assigned functions and either follow designated roles or decide for themselves

what roles to play.32

These abilities of nonhuman animals have rarely been taken seriously in con-

temporary philosophy, and yet they provide plausible reasons to attribute ele-

mentary self-consciousness and some degree of autonomy to nonhuman animals.

Their abilities seem to admit degrees of just the properties that I identified

previously. Any theory similar to the one I presented there must allow for the

possibility (I think inevitability) that some nonhuman animals are at a higher level


of autonomy (or, possibly, personhood) than some humans. Some measure of

autonomy can be gained or lost over time by both humans and some nonhuman

animals as critical capacities are gained, enhanced, or lost.

The fact that humans will generally exhibit higher levels of cognitive capacities

under these criteria than other species of animals is a contingent fact, not a

necessary truth about the human species. A nonhuman animal may overtake a

human whenever the human loses a measure of mental abilities after a cataclysmic

event or a decline of capacity. If, for example, the primate in training in a language

laboratory exceeds the deteriorating human, the primate may achieve a higher

level of autonomy (or, perhaps, personhood), and may thereby be positioned to

gain a higher moral status, depending on the precise connection asserted in the

theory. (Even if animals such as the great apes fail to qualify as autonomous

persons, it does not follow that they have no capacities of autonomous choice.)

Criteria of Morally Autonomous Persons

The category of morally autonomous persons is relatively uncomplicated by

comparison to the category of metaphysically autonomous persons. I will not

attempt an account of the necessary and sufficient conditions, but it seems safe to

assume that X is a morally autonomous person if (1) X is capable of making moral

judgments about the rightness and wrongness of actions and (2) X has motives

that can be judged morally. Being a morally autonomous person, unlike being a

metaphysically autonomous person, is sufficient for moral status. Moral agents

are paradigm bearers of moral status. Any morally autonomous person is a

member of the moral community and qualifies for its benefits, burdens, protec-

tions, and punishments. Nonhuman animals are not, on current evidence, plau-

sible candidates for classification as morally autonomous persons, though some

evidence suggests that the great apes, dolphins, and other animals with similar

properties could turn out to be exceptions. I will not defend this view, but I will

mention a conclusion of Charles Darwin’s that I accept.33 He denied that animals

make moral judgments, but affirmed that some animals display moral emotions

and dispositions. Though they do not make genuine judgments of moral blame

when they punish their peers for misbehavior, they do display genuine love,

affection, and generosity toward their peers. However difficult it is to prove this

thesis, it is no less difficult to disprove it.

Finally, if being morally autonomous is the sole basis of moral status (a view I

do not hold), then many humans lack moral status—and precisely for the reasons

that nonhuman animals do. Fetuses, newborns, psychopaths, severely brain-

damaged patients, and various dementia patients are candidate cases. These

humans are in the same situation as many nonhumans: Moral status for them is


grounded in something other than being morally autonomous. However, this

topic of moral status will have to be the subject of another paper aimed at showing

that certain noncognitive and nonmoral properties, such as emotions and affective

responses, are sufficient to confer some form or level of moral status.

CONCLUSION

Theories of autonomy, like many theories in philosophy, develop from pretheoret-

ical, considered judgments. Considerable vagueness surrounds the ordinary con-

cept of autonomy, and philosophical theories of autonomy that attempt to give the

notion substance should be welcomed. These theories are interesting in their own

right, and practically, they may be of assistance in helping us understand what it is

that we are to respect whenwe respect another’s autonomy. At the same time, in the

development of these theories we should not stray from our pretheoretical judg-

ments about what deserves respect when willed or chosen by another. The moral

value of respect for autonomy precedes and is not the product of a philosophical

theory, and no theory is acceptable if it threatens this value.

Notes

1. Cf. treatments of the concept of autonomy in Joel Feinberg, Harm to Self, Vol. III in

The Moral Limits of the Criminal Law (New York: Oxford University Press, 1986), chaps.

18–19; and Thomas E. Hill, Jr., Autonomy and Self-Respect (Cambridge: Cambridge

University Press, 1991), chaps. 1–4. Individual (by contrast to political) autonomy is a

recent concept lacking a significant history of philosophical analysis. When the eight-

volume Encyclopedia of Philosophy (Macmillan) was published in 1967, it offered no

indexed mention of “autonomy.” (Its sole reference is to “autonomous idiolects”; the

entries under “self-determination” are scarcely more informative.) In its current non-

Kantian uses, the term came into vogue in philosophy shortly after this Encyclopedia

went to press.

2. O’Neill, Rethinking Informed Consent in Bioethics (Cambridge: Cambridge University

Press, 2007), 25, 189. See also her Autonomy and Trust in Bioethics (Cambridge: Cambridge

University Press, 2002), sect. 7.6, 154–160.

3. Julian Savulescu, “Rational Desires and the Limitation of Life-Sustaining Treatment,”

Bioethics 8 (1994): 191–222; and Savulescu and Richard Momeyer, “Should Informed

Consent Be Based on Rational Beliefs?” Journal of Medical Ethics 23 (1997): 282–288. I am

not asserting that Savulescu’s conclusions could never be warranted; obviously a Jehovah’s

Witness could be acting nonautonomously. I am pointing to the profound practical

consequences to which a theory of autonomymay be put and suggesting a different starting

point than the one used by Savulescu.



(New York: Oxford University Press, 1986), chap. 7; Tom L. Beauchamp and James

Childress, Principles of Biomedical Ethics (New York: Oxford University Press, 2001),

chap. 3. Some writers committed to a feminist perspective have argued that my theory is

not entirely compatible with the constraints on theory that I have proposed, or at least that

my theory needs adjustments. See Carolyn Ells, “Shifting the Autonomy Debate to Theory

as Ideology,” Journal of Medicine and Philosophy 26 (2001): 417–430; and Anne Donchin,

“Autonomy and Interdependence,” in Relational Autonomy, ed. C. Mackenzie and N.

Stoljar (New York: Oxford University Press, 2000), 236–258.

5. Alvin I. Goldman, A Theory of Human Action (Englewood Cliffs, NJ: Prentice-Hall,

1970), 49–85.

6. Hector-Neri Casta~neda, “Intensionality and Identity in Human Action and

Philosophical Method,” Nous 13 (1979): 235–260, esp. 255.

7. This analysis draws from Faden and Beauchamp, A History and Theory of Informed

Consent, chap. 7.

8. I follow John Searle in thinking that we cannot reliably distinguish, in many situa-

tions, among acts, effects, consequences, and events. Searle, “The Intentionality of

Intention and Action,” Cognitive Science 4 (1980): esp. 65.

9. This now widely used label seems to have been coined by Paul S. Appelbaum, Loren

Roth, and Charles W. Lidz in “The Therapeutic Misconception: Informed Consent in

Psychiatric Research,” International Journal of Law and Psychiatry 5 (1982): 319–329. For

evidence that the misconception is enduring, see Appelbaum, Lidz, and Thomas Grisso,

“Therapeutic Misconception in Clinical Research: Frequency and Risk Factors,” IRB: Ethics

and Human Research 26 (2004): 1–8. See, further, W. Glannon, “Phase I Oncology Trials:

Why the Therapeutic Misconception Will Not Go Away,” Journal of Medical Ethics 32

(2006): 252–255.

10. See Sam Horng and Christine Grady, “Misunderstanding in Clinical Research:

Distinguishing Therapeutic Misconception, Therapeutic Misestimation, & Therapeutic

Optimism,” IRB: Ethics and Human Research 25 (January–February 2003): 11–16.

11. See the original formulation of this view in Faden and Beauchamp, A History and

Theory of Informed Consent, chap. 10.

12. This formulation is indebted to Robert Nozick, “Coercion,” in Philosophy, Science

and Method: Essays in Honor of Ernest Nagel, ed. Sidney Morgenbesser, Patrick Suppes, and

MortonWhite (New York: St. Martin’s Press, 1969), 440–472; and Bernard Gert, “Coercion

and Freedom,” in Coercion: Nomos XIV, ed. J. Roland Pennock and John W. Chapman

(Chicago: Aldine, Atherton Inc., 1972), 36–37. See also Alan Wertheimer, Coercion

(Princeton, NJ: Princeton University Press, 1987).

13. Dworkin, The Theory and Practice of Autonomy (New York: Cambridge University

Press, 1988), chaps. 1–4; Harry G. Frankfurt, “Freedom of the Will and the Concept of a

Person,” Journal of Philosophy 68 (1971): 5–20, as reprinted in The Importance of What We

Care About (Cambridge: Cambridge University Press, 1988): 11–25. See also Laura W.

Ekstrom, “A Coherence Theory of Autonomy,” Philosophy and Phenomenological

Research 53 (1993): 599–616; and Gary Watson, “Free Agency,” Journal of Philosophy 72


(1975): 205–220. Although it is far from clear that Frankfurt holds a theory of autonomy, see

his uses of the language of “autonomy” in his Necessity, Volition, and Love (Cambridge:

Cambridge University Press, 1999), chaps. 9, 11. Frankfurt’s early work was on persons and

freedom of the will. In his later work, he seems to regard the earlier work as providing an

account of autonomy, which is a reasonable estimate even if it involves some creative

reconstruction.

14. Frankfurt, “Freedom of the Will and the Concept of a Person,” in The Importance of

WhatWe Care About, 16. Frankfurt has, in his later philosophy, modified his early theory of

identification. See his “The Faintest Passion,” in Necessity, Volition, and Love, 95–107, esp.

105–106.

15. Dworkin, The Theory and Practice of Autonomy, 20

16. Frankfurt, “Autonomy, Necessity, and Love,” in Necessity, Volition, and Love, 137.

17. Problems of this sort were first called to my attention by Irving Thalberg,

“Hierarchical Analyses of Unfree Action,” Canadian Journal of Philosophy 8 (1978): 211–226.

18. “The Faintest Passion” and “On the Necessity of Ideals,” in Necessity, Volition, and

Love, 105, 110.

19. “Freedom of the Will and the Concept of a Person,” in The Importance of What We

Care About, 18.

20. “The Faintest Passion,” in Necessity, Volition, and Love, 105–106.

21. There are more demanding theories than these second-order theories. Some theories

require extremely rigorous standards in order to be autonomous or to be persons. For

example, they demand that the autonomous individual be authentic, consistent, indepen-

dent, in command, resistant to control by authorities, and the original source of values,

beliefs, rational desires, and life plans. See Stanley Benn, “Freedom, Autonomy and the

Concept of a Person,” Proceedings of the Aristotelian Society 76 (1976): 123–130, and A Theory

of Freedom (Cambridge: Cambridge University Press, 1988): 3–6, 155f, 175–183; R. S. Downie

and Elizabeth Telfer, “Autonomy,” Philosophy 46 (1971): 296–301; Christopher McMahon,

“Autonomy and Authority,” Philosophy and Public Affairs 16 (1987): 303–328.

22. “The Faintest Passion,” inNecessity, Volition, and Love, 104–105. See esp. the account

of satisfaction. A nuanced version of this theory of nonrepudiated acceptance (using the

language of nonresistance) seems to be held by John Christman, “Autonomy and Personal

History,” Canadian Journal of Philosophy 21 (1991): 1–24, esp. 10ff.


What We Care About, 18.

24. See Savulescu, “Rational Desires and the Limitation of Life-Sustaining Treatment”:

202ff (summarized at 221–222).

25. Other philosophers have used this or a similarly worded distinction, but not as I

analyze the distinction. See Daniel Dennett, “Conditions of Personhood,” In The Identities

of Persons, ed. Amelie O. Rorty (Berkeley: University of California Press, 1976), 175–196, esp.

176–178; Joel Feinberg and Barbara Baum Levenbook, “Abortion,” in Matters of Life and

Death: New Introductory Essays in Moral Philosophy, 3rd edition, ed. Tom Regan (New

York: Random House, 1993), 197–213; Stephen F. Sapontzis, Morals, Reason, and Animals

(Philadelphia: Temple University Press, 1987), 47ff.


26. SeeMichael Tooley, Abortion and Infanticide (Oxford: Clarendon Press, 1983); Harry

Frankfurt, “On the Necessity of Ideals,” in Necessity, Volition, and Love, 113–115, and

“Freedom of the Will and the Concept of a Person,” in The Importance of What We Care

About, ch. 2, esp. 12; Dennett, “Conditions of Personhood”: 177–179; Mary Anne Warren,

Moral Status (Oxford: Oxford University Press, 1997), chap. 1; H. Tristram Engelhardt, Jr.

The Foundations of Bioethics, 2nd edition (New York: Oxford University Press, 1996), chaps.

4, 6; Loren Lomasky, Persons, Rights, and the Moral Community (Oxford: Oxford

University Press, 1987); Lynne Rudder Baker, Persons and Bodies (Cambridge: Cambridge

University Press, 2000), chaps. 4, 6.

27. On the relevance and plausibility of robots and physical-mental systems that imitate

human traits, see John Pollock, How to Build a Person (Cambridge, MA: MIT Press, 1989),

and Ausunio Marras, “Pollock on How to Build a Person,” Dialogue 32 (1993): 595–605.

28. For Kant, a person’s dignity—indeed, “sublimity”—comes from being his or her

own moral lawgiver, that is, from being morally autonomous. Kant, Foundations of the

Metaphysics of Morals, Lewis White Beck, trans. (Indianapolis: Bobbs-Merrill Company,

1959), 58. Kant added that the dignity deriving from this capacity is of a priceless worth that

animals do not have: “[Each person] possesses a dignity (an absolute inner worth) whereby

he exacts the respect of all other rational beings. . . . The humanity in one’s person is the

object of the respect which he can require of every other human being.” Kant, The

Metaphysical Principles of Virtue, pt. I, James W. Ellington, trans., in Kant, Ethical

Philosophy (Indianapolis, IN.: Hackett Publishing Co., 1983), 97–98.

29. As is asserted, more or less, by Harry Frankfurt, “Autonomy, Necessity, and Love,” in

hisNecessity, Volition, and Love, 131, n. 2, and his “Freedom of theWill and the Concept of a

Person,” in The Importance of What We Care About, 16–17; Allen Buchanan and Dan Brock,

Deciding for Others: The Ethics of Surrogate Decision Making (Cambridge: Cambridge

University Press, 1989), 197–199; John Harris, The Value of Life (London: Routledge,

1985), 9–10; Dworkin, The Theory and Practice of Autonomy, esp. chap. 1.


What We Care About, 12.

31. See Donald R. Griffin, Animal Minds (Chicago: University of Chicago Press, 1992);

and Rosemary Rodd, Ethics, Biology, and Animals (Oxford: Clarendon Press, 1990), esp.

chaps. 3, 4, 10.

32. Cf. Gordon G. Gallup, “Self-Recognition in Primates,” American Psychologist 32

(1977): 329–338; David DeGrazia, Taking Animals Seriously: Mental Life and Moral Status

(New York: Cambridge University Press, 1996), esp. 302.

33. Charles Darwin, The Descent of Man and Selection in Relation to Sex (Detroit: Gale

Research Co., 1974), chaps. 3, 4.


6

THE CONCEPT OF PATERNALISM IN

BIOMEDICAL ETHICS

The idea that paternalism has a positive role to play in medicine and public

health descends from ancient ideas about the duties and virtues of physicians.

Traditional notions of the physician’s obligations have been expressed pri-

marily in terms of principles of beneficence and nonmaleficence: “As to

disease, make a habit of two things—to help, or at least to do no harm.”1

Until the recent rise of literature on medical paternalism, medical tradition

placed little significance in a principle of respect for the autonomy of patients.

However, as medicine confronted cultural claims of patients’ rights to make

independent judgments about their medical fate, paternalism loomed ever

larger.

In biomedical ethics today, paternalism has been both defended and

attacked in clinical medicine, public health, health policy, and government

agencies. Unfortunately, it is unclear in much of this literature what patern-

alism is and which types of paternalism, if any, are justified. The closest thing

to consensus in this literature seems to be that so-called strong (or hard)

paternalism is not justified. In this paper I will argue that despite the pre-

ponderance of opinion in the literature on paternalism, so-called strong

paternalism can be justified. I will also argue that it is the only interesting

form of paternalism.

THE IDEA OF AN AUTONOMY-LIMITING PRINCIPLE

In philosophy, the history of criticism of paternalism can be traced to John Stuart

Mill’s attack on paternalism in his theory of a person’s liberty rights and whether

they can be justifiably limited. Although he does not use today’s favored term

autonomy, he defends a personal and community ideal of “the free development

of individuality,” which is the functional equivalent for him of the development of

personal autonomy.2 Mill argues that our liberty rights are sometimes justifiably

overridden because they conflict with other values that we treasure, such as health

and welfare. For example, in a welfare state, people are legitimately taxed against

their will—a deprivation of their liberty to spend their money—in order to benefit

those who need assistance in the form of health care, job insurance, food, child care,

and the like. Such conflicts between the value of liberty and other goods present a

need to establish the proper limits on autonomous choice and action for individuals,

groups, and institutions. Various principles, generally assumed to be moral princi-

ples, have been proposed to justify the limitation of individual human liberties.

In closely tracking Mill’s arguments in On Liberty, Joel Feinberg called these

principles “liberty-limiting principles” and “coercion-legitimizing principles.”3

The issue is the proper moral justification for limiting liberty. I will refer to these

principles as “autonomy-limiting” principles. These principles allegedly state

valid restrictions of an autonomous agent’s liberty. The principles are said by

their proponents to provide a good reason for interference with autonomy and for

any corresponding use of coercion necessary for the interference. The following

are the four most commonly discussed autonomy-limiting principles in the

model proposed by Feinberg:

1. The harm principle: A person’s autonomy is justifiably restricted to pre-

vent harm to others caused by that person.

2. The principle of paternalism: A person’s autonomy is justifiably restricted

to prevent harm to self caused by oneself, irrespective of whether harm is

caused to others.

3. The principle of legal moralism: A person’s autonomy is justifiably

restricted to prevent that person’s immoral behavior, irrespective of

whether that person’s conduct is harmful or offensive.

4. The offense principle: A person’s autonomy is justifiably restricted to

prevent hurt or offense (as contrasted to harm or injury) to others

caused by that person’s behavior.

Mill defended the first principle while rejecting the others. He delivered the

following compelling attack on the principle of paternalism in his monographOn

Liberty:


The only purpose for which power can be rightfully exercised over any

member of a civilized community, against his will, is to prevent harm to

others. His own good, either physical or moral, is not a sufficient warrant. He

cannot rightfully be compelled to do or forbear because it will be better for

him to do so, because it will make him happier, because in the opinions of

others, to do so would be wise, or even right. These are good reasons for

remonstrating with him, or reasoning with him, or persuading him, or

entreating him, but not for compelling him. . . . In the part which merely

concerns himself his independence is, of right, absolute.4

Mill thus regarded true paternalism as always unjustified (and not merely

prima facie unjustified). His arguments, though presented as utilitarian, seem

to rest on additional premises. Purely utilitarian grounds rarely, if ever, warrant

absolute restrictions. Mill’s own formulations of utilitarianism seem insufficient

for the conclusion that even if society can know the best interests of individuals, it

should not advance them by compulsion. Mill seems, in the end, more compelled

by the idea that we have an absolute right not to have our liberty overridden,

whether or not the benefits outweigh the risks.

To move now beyond Mill, paternalism as a philosophical problem arises from

this general theoretical problem of justifying principles that provide grounds for

the limitation of autonomy. There are several distinguishable issues, two being the

chief issues. The first is how we should understand the concept of paternalism.

The second is whether paternalism should be excluded from consideration as a

valid autonomy-limiting principle. I will take them up in turn, starting with how

to formulate the concept of paternalism and the notion of a principle of

paternalism.

THE DEFINITION OF PATERNALISM

In ordinary discourse, by contrast to theoretical discussion of a system of valid

autonomy-limiting principles, the term paternalism dates from the principle and

practice of paternal administration, especially government in the model of

paternal benevolence. The analogy presupposes that a father acts beneficently

toward his children and makes decisions about his children’s welfare without

allowing the child’s preferences to be the determinative consideration. In health

care and public health, this analogy builds on the idea that a professional has

superior training, knowledge, and insight and is thus in an authoritative position

to determine the best interests of those under his or her care and administration.

The nature of paternalism thus begins in analogy, but what, more precisely, is the

meaning of the term paternalism, and how broad is its scope?

T H E C O N C E P T O F P A T E R N A L I S M I N B I O M E D I C A L E T H I C S 103

The Autonomy-Limiting Definition

“Paternalism” could be understood as referring to acts or practices that restrict

the autonomy of individuals without their consent even though they have the

capacity to act autonomously. This conception leads to the following definition,

which I will call the autonomy-limiting definition: Paternalism is the intentional

limitation of the autonomy of one person by another, where the person who limits

autonomy justifies the action exclusively by the goal of helping the person whose

autonomy is limited.5 Following this definition, paternalism seizes decision-

making authority by overriding a person’s autonomous choice on grounds of

providing the person with a benefit—in medicine, a medical benefit. Here an act

of paternalism overrides the principle of respect for autonomy on grounds of the

welfare of the person whose autonomy is overridden.

Examples of paternalism in medicine in this strong sense include court orders

for blood transfusions when patients have refused them, involuntary commit-

ment to institutions for treatment, intervention to stop rational suicides, resusci-

tating patients who have asked not to be resuscitated, withholding medical

information that patients have requested, denial of an innovative therapy to

someone who wishes to try it, and government-enforced efforts to promote

health. Other health-related examples include laws requiring motorcyclists to

wear helmets andmotorists to wear seat belts and the regulations of governmental

agencies that prevent people from purchasing possibly harmful or inefficacious

drugs and chemicals. The motivation in these cases is the beneficent promotion of

health and welfare of those whose autonomous choice is overridden.

The Preference-Limiting Definition

Some writers object to the boldness of autonomy restriction in this autonomy-

limiting definition of paternalism. They judge that it fails to grasp the subtleties of

the term as it has descended from common usage and legal precedent. The notion

is linked in these literatures to guardianship, surrogate decision making, and

government intervention to protect the vulnerable or unwary. The root sense of

paternalism is joined with the law’s wide-ranging use of terms such as parens

patriae to produce a broad meaning of paternalism that includes interventions

with both autonomous actions and questionably autonomous actions. In this

conception, even if a person’s plans or actions do not derive from a substantially

autonomous choice, overriding them can still be paternalistic.6 For example, if a

patient still groggy from surgery attempts to get out of bed, it is paternalistic to

prevent him from doing so despite the fact that he has not made a substantially

autonomous choice.


Those who follow this broader conception recommend the following defini-

tion, which I will call the preference-limiting definition: Paternalism is the inten-

tional overriding of one person’s known preferences by another person, where the

person who overrides justifies the action by the goal of benefiting the person

whose will or preference is overridden. Under this second definition, if a person’s

stated preferences do not derive from a substantially autonomous choice, the act

of overriding his or her preferences is still paternalistic. The main condition of a

paternalistic action is that a beneficent treatment is the reason for overriding a

known preference, whether or not the preference is formed autonomously.7

A Final Definition

These two definitions are both entrenched in current literature, but neither

dominates the literature. I will follow the more limited and far more interesting

commitments of the first definition because of its fidelity to the core problem of

the limitation of autonomy. There are compelling reasons for resisting the second

(preference-limiting) definition on grounds that paternalism originates in ethical

theory as an issue about the valid limitation of autonomy (or liberty), and the

second definition does not require substantial autonomy. To include cases invol-

ving persons who lack substantial autonomy such as drug addicts and the

mentally handicapped broadens the term in a way that obscures the central

issue, which is how, whether, and when autonomy can be justifiably limited.

It might be doubted that the first definition is adequate on grounds that it is not

a necessary condition of paternalism that autonomy be overridden. Perhaps

autonomy is simply not consulted in the sense that a paternalist can act without

caring what the wishes are of the person toward whomhe or she acts beneficently.8

For example, suppose a doctor thinks he knows best and does not consult the

patient’s wishes; he does not know or have any belief about the patient’s choices or

intentions. The doctor simply acts in the patient’s best medical interests. Such an

action is authoritarian and beneficent, but it seems conceptually dubious that

these conditions render the action paternalistic. They conformmore to the idea of

good medical care than they do to medical paternalism. Consider the analogous

case in which a father pays his 21-year-old son’s taxes without consulting his son’s

wishes in the matter. Nothing in this act smacks of paternalism; the father has

always paid his son’s taxes and merely acts in a parental role in his son’s best

interests. In short, unless a specific desire or choice of a person is overridden, an

action falls short of paternalism.

Accordingly, the definition of paternalism I will follow hereafter is this:

Paternalism is the intentional overriding of one person’s autonomous choices

or actions by another person, where the person who overrides justifies the action


by appeal to the goal of benefiting or of preventing or mitigating harm to the

person whose choices or actions are overridden. A person’s choices or actions

must be substantially autonomous to qualify as paternalistic under this definition.

Later in this chapter I will make some qualifications on this notion of “one

person . . . by another” in light of some important issues about the large popula-

tions that confront persons responsible for public policies governing all motorcy-

clists, all institutions that screen for disease, and the like.

My definition is value neutral in the sense that it does not presume whether

paternalism is justified. The term paternalism, following this definition, has

neither a negative nor a positive valence and leaves questions of justification

open. The definition assumes an act of beneficence analogous to parental bene-

ficence, but it does not prejudge whether the beneficent act is justified, obligatory,

or misplaced.

WEAK (SOFT) PATERNALISM AND STRONG (HARD) PATERNALISM

Acceptance of this final definition instead of the broader preference-limiting

definition will play an important role in determining whether one accepts a

type of paternalism often distinguished in the literature as weak paternalism,

also called “soft” paternalism (a distinction owing much to Feinberg).9 I do not

regard weak (or soft) paternalism as a true type of paternalism, but many do so

regard it, and it is important to see what is at stake.

In weak paternalism, a paternalistic actor “has the right to prevent self-

regarding conduct only when it is substantially nonvoluntary or when temporary

intervention is necessary to establish whether it is voluntary or not.”10 Permissible

limitations of autonomy are confined to substantially nonautonomous behaviors,

usually conditions of compromised ability or even dysfunctional incompetence.

Conditions that can significantly compromise the ability to act autonomously

include the influence of psychotropic drugs, painful labor while delivering a child,

and a blow to the head that affects memory and judgment. A physician who

overturns the preferences of a substantially nonautonomous patient in the inter-

ests of medical welfare acts paternalistically and justifiably by the standards of

weak paternalism. Given the reasonableness of the claim that nonautonomous

preferences can be validly overridden, the conclusions of weak paternalism have

been generally accepted in law, medicine, and moral philosophy. However, the

reasonableness of weak paternalistic interventions does not make it an interesting

or conceptually coherent form of paternalism. As Feinberg has pointed out, weak

paternalism is a rather uninteresting type, because it is not clearly a liberty-

limiting principle independent of the harm principle.11 The final definition

I proposed earlier rules it out as a form of paternalism. Moreover, the idea that


paternalism is justified “only when it is substantially nonvoluntary” will be shown

later in this essay to be dubious.

Strong paternalism, by contrast to weak, supports interventions that protect

competent adults by restricting substantially autonomous behaviors. For

example, refusing to release a competent hospital patient who will die outside

the hospital but requests the release knowing the consequences is an act of strong

paternalism. Strong paternalism usurps autonomy by either restricting the infor-

mation available to a person or overriding the person’s informed and voluntary

choices. These choices need not be fully autonomous or voluntary (a notion that

may be empty, perhaps even meaningless), but in order to qualify as strong

paternalism, the choices of the beneficiary of paternalistic intervention must

be substantially autonomous or voluntary. For example, a strong paternalist

might prevent a patient capable of making reasoned judgments from receiving

diagnostic information if the information would lead the patient to a state of

depression and a decision to commit suicide.

Under the final definition I have supported, only strong paternalism qualifies as

paternalism. Accordingly, weak paternalism will hereafter not be under discus-

sion. Whether strong paternalism is justified will of course need close

consideration.

PATERNALISM IN POLICY, PATERNALISTIC INTENT, AND

PATERNALISTIC EFFECT

Those who frame a policy in public policy and public health contexts often seek to

avoid harm to or to benefit persons without directly consulting them. Often it is

known that some percentage of the population would not agree to provisions in

the policy that restrict choice. The policymakers will know that some members of

the group affected regard the policy as autonomy depriving, whereas others

strongly approve of the policy. In effect, the policy is intended to benefit all

members of a population without consulting the autonomous preferences of

individuals, though with the knowledge that many individuals would reject the

control the policy exerts over their lives. Policy contexts, by contrast to clinical

situations in the patient–physician relationship, commonly present this situation.

These policies are paternalistic in intent and commonly will be paternalistic in

their effect on some members of the population, which points to the importance

of a distinction between paternalistic intent and paternalistic effect. A term such

as paternalistic action generally carries the meaning of both intent and effect

(or outcome), but the two should be distinguished. Mere paternalistic intent is

insufficient for paternalistic effect; and it muddies the waters to speak only of

“paternalistic action.” For example, a physician may intend to override a patient’s


choice while misunderstanding that the patient has already chosen what the

physician wants; and a policymaker may formulate a paternalistic health policy,

such as a prohibition of a certain kind of fat in marketed food products, only to

find that no one actually wants that type of fat in food products.

In his work on paternalism, Cass Sunstein has maintained that “The idea of

libertarian paternalism might seem to be an oxymoron, but it is both possible and

desirable for private and public institutions to influence behavior while also

respecting freedom of choice.”12 “Libertarian paternalism” is indeed an oxymoron

on the conceptual account of paternalism I have proposed. However, we can

make sense of, and accept, Sunstein’s proposal by using the distinction between

paternalistic intent and paternalistic effect. Suppose that the available evidence

establishes that smokers psychologically discount the risks of smoking because of

an “optimism bias”; it does not thereby seem correct to say, Sunstein argues, “that

government would violate their autonomy” through programs intended to turn

their biases around—even if these efforts involve manipulative television advertise-

ments that graphically present the suffering that occurs from smoking.13 This claim

is premised on the view that people have a bounded rationality that creates a

limitation on their capacity to act autonomously. Autonomous persons would

value health over the ill health caused by smoking, and in this sense one’s deepest

autonomous commitment is to be a nonsmoker. Accordingly, we are justified on

autonomy grounds in arranging choice situations for populations in a way likely to

correct for cognitive biases and bounded rationality. This strategy favors themanip-

ulation of people into doing what is good for them. So interpreted, Sunstein’s

position is paternalistic (possibly even a hard paternalism), depending on the

nature of the manipulation and the nature of smokers’ choices. The exact measure

of paternalism, if any, will have to be ascertained in the context of each specific case.

Those who make decisions and formulate policies may have a paternalistic

intent, such as that of reversing the autonomous choices of smokers to smoke, and

may act accordingly, but the result may ormay not be paternalism in ameaningful

sense. Suppose the smokers, though they every day choose to smoke cigarettes,

would welcome rather than reject or even resist manipulative efforts to change

behavior. In that event, no autonomy is overridden, and the action taken is not

paternalistic even though a paternalistic intent motivates the action. This patern-

alism presumably reflects values that individuals would recognize or realize

themselves if they did not encounter internal limits of rationality and control.

If such paternalistically motivated plans and policies face no opposition from the

targeted group, they have no paternalistic effect, despite the paternalistic intent.

Some paternalistic policies appear to be based entirely on the value of the

predicted outcomes for the beneficiary, without consideration of whether a

person’s autonomy is diminished and without knowledge of the person’s prefer-

ences. If the intervention is based on evidence that people are not rational decision


makers in the relevant sphere of action, a major part of the appeal is to limited

autonomy or perhaps nonautonomy, in which case it may turn out to be a form of

weak paternalism or no paternalism at all.

THE PROBLEM OF JUSTIFICATION AND THE PRINCIPLE

OF PATERNALISM

The definition or conceptual analysis of paternalism—the center of the discussion

thus far—should be kept distinct from the justification of paternalism. Questions

of justification return us to the philosophical theory deriving from Mill with

which we began: Can a principle of paternalism be justified as an autonomy-

limiting principle, or is it an invalid autonomy-limiting principle? If it is an

invalid principle, then paternalism cannot be justified.

How, then, might strong paternalism be justified? This paternalism might be

justified as a “social insurance policy” that fully rational persons endorse.14 That is,

one might attempt to justify it by the conditions to which an impartial rational

agent would consent if he or she were to appreciate that he or she might be tempted

at times to make decisions to commit acts such as suicide that are potentially

dangerous and irreversible. In this conception a paternalistic act protects the

deepest kind of autonomy by protecting a person’s deepest values and preferences.

A physician might lie to a patient, for example, in order to prevent a suicide, where

the physician knows that the patient at the deeper level wants to live.15

This proposed justification fails to clarify the central issues about the justification

of strong paternalism. Why do we need the indirect device of rational approval?

Would a hypothetically consented-to arrangement even be paternalistic? Most

important, if the paternalist’s true objective is to protect or improve the welfare

of another person, then the intervention finds its justification in the harm avoidance

or benefit production that constitutes welfare promotion, as is the case in the

justification of the root notion of parental actions that override the wishes

of their children. If people should be handled paternalistically, it is not because

they hypothetically would or ultimately will consent or would have consented were

they rational, but because they will be better off. This justification rests on providing

for the welfare of another, not on respecting the other’s autonomy.

The Challenge of Antipaternalism

Some philosophers believe that paternalism is never justified. Themost influential

argument to this effect is that rightful authority resides in the individual because

of the strict demands of the principle of respect for autonomy. Since (hard)


paternalistic interventions necessarily display disrespect toward autonomous

agents and fail to treat them as moral equals, they override (and, some would

say, necessarily violate) the agents’ right to be an independent chooser of their

own good. If others impose their conception of the good, they deny the respect

owed us, even if they have a better conception of our needs than we do and act to

benefit us. In effect, the argument is that individuals have a right of autonomy that

is absolute and can never be overridden on grounds merely of welfare interests.

Mill often seems to embrace such a strident antipaternalism, despite the fact

that his philosophy (utilitarianism) is centrally one of balancing different

welfare interests so as to maximize overall value. This theory seems to reach

directly out to an embrace of paternalism, but this is not Mill’s view. He

maintains only that we can permissibly restrict a person’s autonomy tempora-

rily to ensure that the person is acting autonomously, and therefore acting with

adequate knowledge of the consequences of his or her action. Once warned, the

person must be allowed, Mill says, to choose whatever course he or she desires.

Like Mill, virtually all antipaternalists permit a temporary infringement of

autonomy in the belief that persons who have a well-formed, autonomous

resolution to do something harmful to themselves will eventually have an

opportunity to perform the action after the temporary intervention. This posi-

tion is akin to an appeal to a precautionary principle that provides an oppor-

tunity to determine whether an action would be paternalistic. It justifies

caution, not paternalism.

However, in a theory that is more strongly committed to balancing benefits,

risks, and values such as self-development, on what conceivable grounds could

paternalism be prohibited? The only grounds would be that, on balance,

autonomy always outweighs all other interests. But what argument would

anchor such a position? Perhaps a theory of rights might, but rights theories are

not usually so grounded. Such theories in the end will face the same problem: An

individual’s rights will have to be balanced against the rights of others and also

against a variety of interests such as public welfare. Defense of absolute autonomy

rights is an unpromising possibility.

The Justification of Paternalism

The main question about the justification of paternalism, then, is whether con-

ditions can be specified that are reasonable and allow for some range of justifiable

limitations of autonomy without opening up a heavy medical paternalism or state

paternalism. A paternalist might propose various conditions with the goal of

keeping paternalism to a minimum, yet without absolutizing antipaternalism. For


example, the paternalist might maintain that interventions are justified only if

there are circumstances in which no acceptable alternative to the paternalistic

action exists, a person is at risk of serious harm, risks to the person introduced by

the paternalistic action itself are not substantial, projected benefits to the person

outweigh risks to the person, and any infringement of the principle of respect for

autonomy is minimal. The thesis is that these conditions are necessary conditions

of any justified paternalistic action.

The most plausible justification of paternalistic actions places benefit on a

scale with autonomy interests and balances both: As a person’s interests in

autonomy increase and the benefits for that person decrease, the justification

of paternalistic action becomes less plausible. Conversely, as the benefits for a

person increase and that person’s autonomy interests decrease, the justification

of paternalistic action becomes more plausible. From this perspective, pre-

venting minor harms or providing minor benefits while deeply disrespecting

autonomy will lack plausible justification, but actions that prevent major harms

or provide major benefits while only trivially disrespecting autonomy may be

justified paternalistic actions.

The claim is that as the risk to a patient’s welfare increases or the likelihood of

an irreversible harm increases, the likelihood of a justified paternalistic inter-

vention correspondingly increases.16 For example, if a dying patient who will live

for one month in the most extreme form of pain refuses pain medication on

grounds that he might become addicted to the medication, it is plausibly

justified for a physician to administer an undetectable pain medication without

informing the patient of the decision. It would not be difficult to generate a large

body of cases in which benefits are major and risks minor (even effectively zero)

and it is appropriate to act with a minor autonomy offense. Such hypothetical

cases might seem contrived and less than convincing. Accordingly, in the

remainder of this article I propose to test the hypothesis of a justified patern-

alism by considering a range of practical problems in medicine and public

policy. Justified paternalism can be found in some of these cases, whereas in

others it cannot.

THE PLACE OF PATERNALISM IN BIOMEDICAL ETHICS

I will consider seven common problems of paternalism in biomedical ethics, some

of which involve policies and laws and some of which do not. In each I will

indicate the values that need to be balanced when making a judgment about

whether a paternalistic intervention is justified.


Reporting and Preventing Suicide Attempts

Many policies and laws about reporting, preventing, or intervening in suicide are

paternalistic—either paternalistic in intent or in effect, or both. It has often been

said that a principle of respect for life creates an obligation to prevent suicide that

overrides obligations of respect for autonomy. A weaker account relies on Mill’s

strategy of temporary paternalism: Intervention is justified to establish autonomy

in the person, but after it is determined that the person’s decisions are substan-

tially autonomous, further intervention would be unjustified. However, the most

straightforward account of laws and policies that allow or promote suicide

intervention is the following paternalistic thesis: Preventing suicide is justified

by the goal of preventing a person from doing serious self-harm. This rationale is

grounded in the view that other persons do sometimes know our best interests

with more insight and foresight than we do, even when we act autonomously.

A related view that supports paternalism in suicide intervention is that friends

and health care professionals are relieved of (i.e., excused from) duties of con-

fidentiality when they report suicide threats. Some writers and policymakers even

defend a paternalistic obligation to report suicide threats.17

Involuntary Hospitalization of Persons Harmful to Themselves

A vast literature surrounds the involuntary hospitalization of persons who have

never harmed others or themselves, but now stand in danger of harming themselves

(or at least are vulnerable to harms that might occur to them, as is not infrequently

the case with homeless persons). Part of the contemporary rationale for use of

police powers for the emergency detention and civil commitment of those dan-

gerous to themselves is a paternalism supported by the knowledge that treatment

has often helped persons over a momentary crisis. These interventions can involve a

double paternalism: a paternalistic justification for commitment and a paternalistic

justification for forced therapy (e.g., psychotherapy) after commitment.

On the one hand, it is difficult to justify the view that institutionalization

involves only a minor harm. On the other hand, if the benefit were sufficiently

large, it might, on balance, outweigh the harm done, especially if institutionaliza-

tion is of a short duration. Consider the classic case of Catherine Lake, who

suffered from arteriosclerosis, which caused temporary confusion andmild loss of

memory, interspersed with times of mental alertness and rationality. All parties

agreed that Lake never harmed anyone or presented a threat of danger. Still, she

was committed to a mental institution because she often seemed in a confused

and defenseless state as she walked city streets. Her condition was interspersed

with periods of mental alertness and rationality in which she clearly could make


autonomous judgments. At a court trial, while rational and capable of making

autonomous choices, she testified that she knew the risks of living outside the

hospital and preferred to assume those risks rather than remain in the hospital.

The court of appeals denied her petition, arguing that she was “mentally ill,”

a “danger to herself,” and “not competent to care for herself.” In its legal

justification, the court cited a statute that “provides for involuntary hospitaliza-

tion of a person who is ‘mentally ill and, because of that illness is likely to injure

himself.’ ”18 The court interpreted the law as having paternalistic intent, and in

Lake’s case it also had a paternalistic effect.

Antipaternalists argue that since Lake did not harm others and understood the

dangers under which she placed herself, her freedom should not have been

restricted. However, this case could and probably should be viewed very differ-

ently if there were a strong promise that Lake would be cured during the course of

a brief hospitalization, despite her resistance to the hospitalization. From this

perspective, a paternalistic intervention is justified by the intent to benefit and the

likelihood of benefit, which outweighs obligations to respect autonomous choice.

The Risks and Benefits of Treatment Refusal

Should a beneficial medical procedure be withheld or withdrawn because the

patient autonomously refuses the procedure? The right to refuse treatment is

often treated in the literature of biomedical ethics as absolute, that is, as a right

that cannot legitimately be balanced against other considerations. Justifications

for overruling a patient’s refusal need not be paternalistic, but they often are

paternalistic because their objective is to prevent harm that would be caused by

the patient’s refusal. The issue is not whether a physician in fact knows what is best

for the patient, but whether the patient has a right to refuse treatment irrespective

of what the physician knows. The patients here envisioned are competent to refuse

therapy, because otherwise the act would presumably not be paternalistic.

A commonplace example of justified paternalism that overrides an autono-

mous refusal appears in the following case:

After receiving his preoperative medicine, C, a 23-year-old male athlete

scheduled for a hernia repair, states that he does not want the side rails up

[on his bed]. C is of clear mind and understands why the rule is required;

however, C does not feel the rule should apply to him because he is not the

least bit drowsy from the preoperative medication and he has no intention of

falling out of bed. After considerable discussion between the nurse and

patient, the nurse responsible for C’s care puts the side rails up. Her

justification is as follows: C is not drowsy because he has just received the


preoperative medication, and its effects have not occurred. Furthermore, if

he follows the typical pattern of patients receiving this medication in this

dosage, he will become drowsy very quickly. A drowsy patient is at risk for a

fall. Since there is no family at the hospital to remain with the patient, and

since the nurses on the unit are exceptionally busy, no one can constantly

stay with C to monitor his level of alertness. Under these circumstances the

patient must be protected from the potential harm of a fall, despite the fact

that he does not want this protection. . . . The nurse restricted this

autonomous patient’s liberty based on . . . protection of the patient from

potential harm . . . and not as a hedge against liability or for protection from

criticism.19

Medical interventions of this sort present minor risks and offer substantial

protection. Their commonness in hospitals is by itself no justification of such

actions, but the lowered risk makes such actions plausible candidates for status as

justified in the circumstances.

Withholding Information

Physicians and families sometimes maintain that a potentially devastating diag-

nosis or prognosis should not be disclosed to a patient. The concern is that bad

news will adversely affect the patient’s health or lead the patient to commit

suicide. If the patient asks for the information or expects a truthful disclosure, it

is paternalistic to withhold the truth, and it is also paternalistic to not disclose

the information if the patient has a right to the information. Similarly, genetic

counselors who use potential marital conflict for a patient as a reason not to

disclose a condition such as nonpaternity deprive a patient of information that

was itself generated in part by materials the patient provided.

In one case a physician obtains the results of a myelogram (a graph of the

spinal region) following examination of a patient. The test yields inconclu-

sive results and needs to be repeated, but also suggests a serious pathology.

When the patient asks for detailed information, the physician decides to

withhold potentially negative information, knowing that the patient would

be highly distressed by the information. The physician is also confident that

nondisclosure will not affect the patient’s consent to another myelogram.

However, the physician is committed to being completely truthful with the

patient in the future about the results of the second test and about needed

procedures. This physician’s act of nondisclosure seems warranted even

though paternalistic.20


Government Paternalism in Health Policy

Some government bureaus function in ways that follow a model of paternalistic

guardianship. They are illustrative of acting with paternalistic intent, though

sometimes with little or no paternalistic effect on the targeted population. The

Food and Drug Administration (FDA) in the United States is an example, because

the agency was chartered to restrict persons from purchasing foods, drugs, and

medical devices that are unsafe or inefficacious even if they wish to purchase

them. A controversial decision by the FDA in 1992 to severely restrict the use of

silicone-gel breast implants exemplifies paternalistic controversies that have beset

the FDA. Women had elected to use implants for over 30 years, either to augment

their breast size or to reconstruct their breasts following mastectomies. Over 2

million women in the United States had had the implants (3 million worldwide)

when, in April 1992, the FDA restricted the use of silicone-gel breast implants until

additional studies could be conducted. Concerns centered on the implants’ long-

evity, rate of rupture, and causal link to various diseases. Those who defended this

restriction contended that no woman should be allowed to take a risk of

unknown, but potentially serious, magnitude, because her consent could not be

informed. The FDA defended a restrictive policy, rather than prohibition, because

it held that patients with breast cancer and others have a legitimate need for breast

reconstruction. The FDA distinguished sharply between reconstruction candi-

dates and augmentation candidates, arguing that the favorable risk–benefit ratio

is confined to reconstruction candidates.21

This decision has deservedly been criticized as inappropriately paternalistic.

The sweeping character of the decision suggests that the government crossed the

line into an unjustified paternalism by overreacting to what were probably

overstated risks, and there had been inadequate consultation with those who

would experience the risks. The government failed to duly consider whether the

subjective benefits for many women outweigh the identified risks. At the time

opinion surveys indicated that most women receiving the implants were satisfied

with them. In light of this result, the most defensible policy would seem to be to

permit use of silicone-gel breast implants while demanding that adequate disclosure

be provided about risks. Raising the level of disclosure standards is usually a more

appropriate approach than raising the level of paternalistic restraints on choice.

It does not follow that the only adequate policy is one of better disclosure or

that paternalism is never justified in such a case. If there were a targetable group of

particularly vulnerable women with high risk of harm from silicone implants, a

paternalistic prohibition might well be an adequate policy for this particular

group. As with other cases, the best policy is one that carefully weighs the benefits

and risks, and then weighs the result against patients’ rights to information

and choice.


Risk in the Formulation of Health Policy

Paternalism in health policy has played a well-known role in the prohibition or

punishment of smoking, drinking, and hazardous recreational activities such as

hang gliding, mountain climbing, and whitewater rafting. Extreme paternalistic

prohibitions are usually avoided in health policy, but minor-risk paternalistic

interventions are common in these policies. Two classic cases in American law

involve requirements that helmets be worn when riding motorcycles. In one case,

a judge relied extensively on Mill’s views about autonomy limitation. He argued

that requiring motorcyclists and their passengers to wear crash helmets for

paternalistic reasons was an instance of reasoning that could lead to unlimited

state paternalism. He acknowledged that highway safety is a relevant reason for

legal restrictions, but not where the cyclist alone is at risk. This legal opinion reads

almost as if it were a direct application of Mill’s philosophy.22 By contrast, a judge

in a different case argued that “we ought to admit frankly that the purpose of the

helmet [requirement] is to preserve the life and health of the cyclist”—and not to

promote public health or highway safety. The judge saw justifiable paternalism.23

This judgment is reasonable. The cost for the cyclist is minimal, and the potential

benefit for the cyclist is life as a healthy person. Only a legal system of absolute

rights would force us to a different conclusion.

Research Ethics: The Use of Prisoners and Vulnerable Groups in Research

Problems in research ethics only infrequently raise issues of paternalism, but there

are exceptional cases. An example appears in biomedical research involving

prisoners. This case is interesting for its illustration of how an action may

appear to be paternalistic when in fact it is nonpaternalistic, but could easily

have been paternalistic.

In its report on research involving prisoners, the National Commission for the

Protection of Human Subjects of Biomedical and Behavioral Research argued that

the closed nature of prison environments creates a potential for abuse of authority

and therefore invites the exploitation and coercion of prisoners.24 Although a

study authorized by the National Commission indicated that most prisoners

wanted to accept the research opportunity and believed that neither coercion

nor undue influence would compromise their consent to research, the National

Commission argued that prisons’ coercive and exploitative possibilities justify

regulations prohibiting the use of prisoners in research, even if they volunteer to

participate.

This restriction at first sight seems paternalistic, but closer analysis in this

particular case shows that it was not presented paternalistically in the Natonal


Commission’s report. The National Commission used a hypothetical to argue

that if a prison environment were not exploitative or coercive (and if a few other

conditions were met), then prisoners should be allowed to choose to participate in

research. The National Commission’s justifying ground was that we cannot

predict or control whether prisoners will be exploited in settings that render

them vulnerable, and that society should prohibit even research to which pris-

oners might validly consent in some prisons because society cannot adequately

monitor whether the subjects’ consent is, in fact, valid and what the subsequent

conditions of participation will be.

A variety of circumstances in biomedical ethics suggest a need to examine

whether so-called vulnerable classes of persons, including prisoners, are being

exploited. Restrictions placed on members of these classes may or may not be

paternalistic. For example, healthy, nonrelated organ donors and cancer

patients solicited for research are sometimes treated in a manner similar to the

prisoners just mentioned. Protective requirements aimed at the protection of

their welfare may or may not be paternalistic. There is no reason why a balance

of considerations might not justify a paternalistic policy in these cases. In some

cases the justification of an act, policy, or practice of nonacquiescence or

intervention in a person’s preferences may be partially, but not purely, pater-

nalistic because it is intermixed with nonpaternalistic reasons, such as protec-

tion of third parties. Such impure or mixed paternalism is common in public

policy debates.

CONCLUSION

Many paternalistic interventions are unjustified. It does not follow that they all

are. I have argued that it is an open question whether reasonably minor offenses to

autonomy, such as withholding certain forms of information, are justified in light

of critical medical goals, such as the administration of lifesaving therapies and the

prevention of suicide. It will well serve moral reflection if we do not follow Mill in

the banishment of a principle of strong paternalism.

I have not argued that it is part of the justification of medical paternalism that

physicians are explicitly authorized to take care of patients. However, the fact

that physicians are positioned to make sound and caring decisions from a

position of authority merits special consideration when judging whether

paternalism in medicine is warranted. In biomedical ethics we are accustomed

to thinking of the right to authorize as belonging exclusively to patients and

their surrogates. This way of thinking about the conduct of physicians is short-

sighted.


Notes

1. Epidemics 1, 11, in: Jones, W. H. S. (ed.) (1923): Hippocrates, I, Cambridge, MA:

Harvard University Press, 165.

2. Mill, J. S. (1977):On Liberty, 4th ed., in: Collected Works of John Stuart Mill, vol. XVIII,

Toronto: University of Toronto Press, 261.

3. See Feinberg (1986-v4) in his four-volume work TheMoral Limits of the Criminal Law,

principally the final volume, Harmless Wrongdoing, New York: Oxford University Press, 9.

The schema of principles outlined here is heavily indebted to Feinberg’s work.

4. Mill, J. S (1977): On Liberty, 223.

5. Dworkin, G. (1992): “Paternalism,” in: Becker, L. (ed.): Encyclopedia of Ethics,

New York: Garland Publishing, 939–942; Arneson, R. (1980): “Mill versus Paternalism,”

Ethics 90, 470–489; Archard, D. (1990): “Paternalism Defined,” Analysis 50, 36–42;

Beauchamp, T.L., McCullough, L.B. (1984): Medical Ethics: The Moral Responsibilities of

Physicians, Englewood Cliffs, NJ: Prentice-Hall.

6. See VanDeVeer, D. (1986): Paternalistic Intervention: The Moral Bounds on

Benevolence, Princeton, NJ: Princeton University Press, 16–40; Kleinig, J. (1983):

Paternalism, Totowa, NJ: Rowman and Allanheld, 6–14.

7. VanDeVeer, D. (1986); Kleinig, J. (1983).

8. Feinberg (1986-v3),Harm to Self (in the four-volume work on The Moral Limits of the

Criminal Law), 5; Seana Shiffrin (2000): “Paternalism, Unconscionability Doctrine, &

Accommodation,” Philosophy and Public Affairs 29, 205–251.

9. Feinberg (1986-v3).

10. Feinberg, J. (1971): “Legal Paternalism,” in:Canadian Journal of Philosophy 1, 105–124,

esp. 113, 116. Revised (1973) in Social Philosophy, Englewood Cliffs, NJ: Prentice-Hall; and

later restated in Harm to Self, Chapter 17.

11. Feinberg, (1971): “Legal Paternalism,” 113, 124. Though some say that weak patern-

alism is the only form of justified paternalism, I believe that it is easy to justify the

conclusions of weak paternalism independent of any mention of a principle of paternalism.

Everyone supports altruistic beneficence directed at confused cardiac patients, ignorant

consumers, frightened clients, and young persons who know little about the dangers of

alcohol, smoking, drugs, and motorcycles. No caring person would leave these individuals

unprotected, and no reasonable philosopher would defend a normative thesis that permits

such outcomes. The knotty problems about the justification of paternalism lie not in these

behaviors. They lie exclusively in strong paternalism, which takes over and overrides

autonomy.

12. Cass R. Sunstein and Richard H. Thaler, “Libertarian Paternalism is Not an

Oxymoron,” University of Chicago Law Review 70 (2003): 1159–1202, esp. 1159.

13. Christine Jolls and Cass R. Sunstein, “Debiasing through Law,” The Journal of Legal

Studies 33 (2006): 232.

14. Dworkin, G. (1972): “Paternalism,” in: The Monist 56, 64–84, esp. 65.

15. Dworkin (1972); see also VanDeVeer (1986) and Kleinig (1983).

16. Compare Kleinig’s similar conclusion in Paternalism, 76.


17. Bloch, K.E. (1987): “The Role of Law in Suicide Prevention: Beyond Civil

Commitment—A Bystander Duty to Report Suicide Threats,” in: Stanford Law Review

39, 929–953.

18. See Katz, J., Goldstein, J., and Dershowitz, A.M. (eds.) (1967): Psychoanalysis,

Psychiatry, and the Law. New York: The Free Press, 552–554, 710–713; and Burt, R.A.

(1979): Taking Care of Strangers: The Rule of Law in Doctor-Patient Relations, New York:

The Free Press, chap. 2.

19. Silva, M.C. (1989): Ethical Decision-making in Nursing Administration, Norwalk, CT:

Appleton and Lange, 64.

20. Beauchamp, T.L., and Childress, J.F. (2009): Principles of Biomedical Ethics, 6th ed.,

New York: Oxford University Press, 215.

21. Kessler, D.A. (1992): “Special Report: The Basis of the FDA’s Decision on Breast

Implants,” in: New England Journal of Medicine 326, 1713–1715.

22. American Motorcycle Association v. Department of State Police 158 N.W. 2d 72. Mich.

App (1968).

23. Florida v. Eitel 227 So. 2d 489 (Fla. 1969).


Behavioral Research (1976): Report and Recommendations: Research Involving Prisoners,

Washington, DC: DHEW Publication No. OS 76-131.


7

WHEN HASTENED DEATH IS NEITHER KILLING

NOR LETTING DIE

There are two fundamental moral issues about physician-assisted hastening of

death: Are physicians morally justified in complying with requests by patients

who ask for assistance? Is there an adequate moral basis to justify the legalization

of physician-assisted hastening of death at the patient’s request? Legal develop-

ments in the United States have encouraged us to frame virtually all moral

questions as ones of legalization. As important as legalization is, the question of

whether physicians are morally justified in complying with requests for aid in

dying presents the more fundamental moral issue. Accordingly, my concern is

primarily with the moral justification of individual acts of hastening death.

The thesis that physician-assisted hastening of death is morally prohibited has

commonly been grounded in the premises that there is a defensible and relevant

distinction between letting die and killing and that physicians may let patients die

but cannot kill them. I do not reject the distinction between killing and letting die;

indeed, I offer an analysis that makes sense of it. Nonetheless, this distinction is

fatally deficient and flawed as a means of treating the major problems of physician-

assisted death that face medicine and society today. Physician-assisted hastening of

death need not involve either killing or letting die; and even if an act is one of killing

or letting die, it must be shown whether the act is justified or unjustified. The

categories of “killing” and “letting die” therefore need to be redirected to the key

moral issue that should drive discussions of physician-assisted death, which is the

liberty to choose and the justification (if any) for limiting that liberty.1

THE CONCEPTUAL FOUNDATION OF THE DISTINCTION BETWEEN

KILLING AND LETTING DIE

In examining the conceptual foundation of the distinction between killing and

letting die, the first question is whether physicians kill patients by causing their

deaths either through interventions or through intentional noninterventions. In

medical tradition, the term killing has understandably carried meanings of wrong-

fulness and blameworthiness. However, in contexts external to traditional med-

ical morality, “killing” does not imply wrongful behavior; it refers only to causal

action that brings about another’s death. In settings both within and outside

medicine, “letting die” refers to intentionally not intervening so that disease,

system failure, injury, or circumstance causes death.

In ordinary discourse, medicine, and law, these conceptions are conspicuously

vague. Some cases of letting die count also as acts of killing, thereby undermining

the hypothesis that these terms distinguish two different sets of cases. For

example, health professionals kill patients—that is, cause patients to die—when

they intentionally let them die in circumstances in which a duty exists to keep the

patients alive, such as a patient with pneumonia who could easily be cured and

wants to be cured but is “allowed to die” by a physician who chooses not to

“intervene.” In much of the literature of medical ethics, it is unclear how to

distinguish killing from letting die so as to avoid even this elementary problem of

cases that satisfy the conditions of both categories.

A widely accepted account of letting die holds that intentionally forgoing a

medical technology qualifies as letting die, rather than killing, only if an under-

lying disease or injury is intentionally allowed to cause death when death might be

delayed by employing that technology.2 According to this view, if a medical

technology is intentionally withdrawn or withheld, a natural death occurs when

natural conditions do what they would have done had the technology never been

initiated (i.e., when existing conditions take their natural, undeterred course). By

contrast, killing occurs if an act of a person or persons, rather than natural

conditions, causes an ensuing death.

This account misses its mark unless other conditions are added to it, but

additional conditions threaten to undermine the very point of the account. One

condition that must be added is that any withholding or withdrawing of a medical

technology must be justified; an unjustified withholding or withdrawing that

releases natural conditions is a killing. This condition brings a moral considera-

tion of justified action into the heart of the conceptual analysis of letting die. To

see why this condition of justified withholding or withdrawing must be added,

consider the case of a physician who either mistakenly (through negligence) or

maleficently (through ill intent) removes a life-sustaining medical technology

WH E N H A S T E N E D D E A T H I S N E I T H E R K I L L I N G N O R L E T T I N G D I E 121

from a patient who wants to continue living. This action lacks justification. We

could not reasonably say, “The physician did not cause the patient’s death; he only

allowed the patient to die of an underlying condition.” In this case, the physician

did cause the death, and the physician did “let the patient die” in the sense

normally meant in medical morality.

Now, change the patient’s wishes. Suppose that the patient autonomously

refused the technology that sustains life. In this circumstance, it would not be

correct to say that the physician caused the death. Whether the physician did or

did not cause death is determined in these circumstances by whether the act was

validly authorized. If the act is validly authorized, it is a letting die; if it is not

validly authorized, it is a killing. The reason for forgoing a medical technology is

therefore the key condition in conceptually distinguishing killing and letting

die in cases of withholding or withdrawing life-sustaining interventions. A phy-

sician lets a patient die if he or she has valid authorization for withholding or

withdrawing treatment. “Letting die” is validly authorized nonintervention in

circumstances in which patients die (when and in the precise manner they do) as a

result. By contrast, a comparable action or inaction is a killing if:

� the physician had a duty to treat,� the physician withheld or withdrew a life-sustaining technology without

authorization, and� the patient subsequently died for lack of that technology.

Here a physician is the relevant cause of death and kills a patient if he or she has

no valid authorization for withholding or withdrawing treatment, but the physi-

cian lets the patient die if he or she does have valid authorization for withholding

or withdrawing treatment. Of course, physicians also may kill in so-called active

ways. I confront this problem in due course.

LETTING DIE BASED ON MEDICAL FUTILITY

Thus far I have argued that the following are conceptually (rather than causally)

sufficient conditions of letting die in medicine:

1. the patient validly refused a medical technology that is essential to the

patient’s continued existence,

2. the physician withheld or withdrew the refused life-sustaining tech-

nology, and

3. the patient subsequently died for lack of the technology.


To say that these conditions are conceptually sufficient for letting die is not to

say that they are either necessary conditions or the only set of sufficient condi-

tions. Withdrawing or withholding a medically futile technology can be the main

reason why we categorize an action as one of letting die. Accordingly, the

following conditions form a second set of (conceptually) sufficient conditions

of letting die in medicine:

1. a medical authority appropriately judged that a medical technology is

futile to achieve the goals for which it was initiated (though it has to this

point kept the patient alive),

2. the physician withheld or withdrew the futile technology, and

3. the patient subsequently died for lack of the technology.

A common, but incorrect, thesis is that letting die occurs in medicine only if

“ceasing useless medical technologies” eventuates in the patient’s death.3 This

account rightly connects letting die to futility (as in 1’ through 3’) but wrongly

assumes that in the circumstance of letting die, technologies must be futile.

As demonstrated in the previous section, medical futility is not a necessary

condition of letting die. A patient’s valid refusal of a medical technology makes

a circumstance one of letting die even if the technology is not futile.

In the case of 1’ through 3’, the physician’s intention may also need to be of

one type rather than another to qualify as letting die, but I will not here pursue

this question about proper intention. What does need an additional comment

is the language in 1’ of “a medical authority appropriately judged that.” As the

previous analysis suggests, in the medical context letting die is conceptually

tied to acceptable acts, where acceptability derives either from a well-substan-

tiated judgment of the futility of a technology or from a valid refusal of the

technology. In the case of a valid refusal, there exists no problem about

responsibility for the death because the refusal itself justifies the physician’s

conduct; the refusal nullifies what would otherwise be an injury or maltreat-

ment and makes the case one of letting die. A judgment of futility does not

as transparently provide justification, but a well-substantiated judgment of

futility can be the basis of a justified act of withholding or withdrawing a

technology.

The language of letting die is used in medical contexts to express a moral

judgment. A letting die is a justified act; it is not morally neutral or unjustified.

Despite this general feature of medical discourse, it is an open question whether

physicians kill or let die when they nonnegligently make a mistaken judgment

about a patient’s condition (after which the patient refuses treatment) or about


the futility of a treatment. Assessment of such hard cases is likely to turn on special

features of the cases. For present purposes, suffice it to say that a conscientious

and knowledgeable physician whomakes a reasoned and justifiable determination

of futility in light of all the information that could be gained in the circumstances

lets the patient die and does not kill the patient, even if the judgment of futility

turns out to have been mistaken.

DO PHYSICIANS KILL PATIENTS WHEN THEY ASSIST

IN HASTENING DEATH?

A necessary condition of killing a patient is that the actions or inactions of the

physician cause the patient’s death. However, the criteria of causing death in the

medical setting need clarification. The conscientious and informed physician

who either makes a justifiable determination of futility or follows a patient’s

valid refusal does not cause the patient’s death even if, in the circumstances, the

technology withdrawn or withheld is causally necessary for the patient’s con-

tinued existence. The physician’s act of withholding or withdrawing is a neces-

sary (or “but for”) condition of the patient’s death as the death occurred, but the

physician is not the cause in the pertinent sense, which is that of being causally

responsible. A physician is not responsible for the consequences of withholding

or withdrawing a technology the patient has refused, even if the physician is a

contributing cause of those consequences. By contrast, the previously men-

tioned physician whomaleficently removed a medical technology from a patient

who wanted to continue living is the cause of death, is causally responsible for the

outcome, and indeed killed the patient.

Why is the first physician not the cause, whereas the second is the cause? “The

cause” judgments (or singular causal judgments) are relative to the prevailing

criteria in a given context of investigation. In the circumstance of physician-

caused death, judgments of causation (and, derivatively, of killing) turn on

whether a physician intervened in an unwarranted manner in a course of events

that could reasonably have been expected to take place. When the physician has

specific warrant for action or inaction (e.g., when the physician has a valid

authorization from a patient) and the patient’s death is a consequence of acting

or not acting, the physician is not the cause of death. Rather, the cause is disease,

injury, system failure, or perhaps the decision of another party who authorized the

physician’s conduct.

To illustrate these abstract points and make them more transparent, consider

the following thought experiment: Two patients occupy a semiprivate hospital


room, both having the same illness and both respirator dependent. One has

refused the respirator; the other wishes to remain on the respirator. A physician

intentionally flips a master switch that turns off both respirators; the physician is

aware that both respirators will shut down. The two patients die in the same way

at the same time of the same physical causes and by the same physical action of the

physician. (Thus, all variables are held constant except that one patient authorized

the physician’s action and the other did not.) Although the two patients die of

the same physical causes, they do not die of the same causes of interest to law,

medicine, and morality, because the proximate cause—that is, the cause respon-

sible for the outcome—is not the same in the two otherwise identically situated

patients. Consistent with the analysis provided previously, this thought experi-

ment shows that a valid authorization transforms what would be a maleficent act

of killing into a nonmaleficent act of letting die.

From both a legal and a moral point of view, one reason why physicians do not

injure, maltreat, or kill patients when they withhold or withdraw medical tech-

nology is that a physician is morally and legally obligated to respect a valid refusal.

Since a valid refusal of treatment binds the physician, it would be unacceptable to

hold that these legal and moral duties require physicians to cause the deaths of

their patients and thereby to kill them.

Quite apart from killing by means of withholding or withdrawing a treatment,

killing can and occasionally does occur through a physician-initiated “active”

means to death. Here is a conceptually clear case: Paul Mills was a patient in the

Queen Elizabeth II Health Sciences Center in Halifax, Nova Scotia. Mills had

undergone 10 unsuccessful operations for throat cancer. He was dying, and a

life-support system had been withdrawn at the request of his family. It was

thought that he would die a natural death within a few hours. However, the

heavy sedation he had been given was not having its intended effect. Mills was

suffering from infection and experiencing “tremendous discomfort” and

“excruciating pain,” according to hospital officials. On November 10, 1996,

Nancy Morrison, Mills’s physician, administered to Mills a dose of potassium

chloride. Mills died shortly thereafter. His family was unaware of the injection.

On May 6, 1997, Dr. Morrison was indicted for first-degree murder; these

charges were thrown out on February 27, 1998, for lack of legal evidence

sufficient to sustain the charge of first-degree murder.4

Dr. Morrison did not, and would not, describe her act as “killing,” let alone as

“murder.” She viewed it as a compassionate act of assistance in dying. It may

indeed have been an act of this description. However, Dr. Morrison had no

specific authorization for her act of injecting potassium chloride, and she was

the cause of Mills’s death in the way and at the time that his death occurred; she

was causally responsible for the death, and she killedMills. Whether she killed him

justifiably is another matter, and one I will now consider.

W H E N H A S T E N E D D E A T H I S N E I T H E R K I L L I N G N O R L E T T I N G D I E 125

IS KILLING BY PHYSICIANS MORALLY PERMISSIBLE?

I have proposed that in cases of killing a patient the physician is causally respon-

sible (a proximate cause), and in cases of letting die the physician is not causally

responsible (not the proximate cause). This proposal so far conforms to legal and

medical traditions. Now, however, I depart from those traditions.

It does not follow from the fact that a physician kills a patient that the physician

acts unjustifiably. Outside of traditional thinking about medical morality, it is

clear that to correctly apply the label “killing” or the label “letting die” to a set of

events cannot determine whether one type of action is better or worse than the

other or whether either is acceptable or unacceptable.5 Rightness and wrongness

in killing depend exclusively on the merit of the justification underlying an act of

killing, not on the type of action it is.

Outside of (and independent of) medical tradition, there are several generally

accepted justifications for killing, including killing in self-defense, killing to rescue

a person endangered by another person’s immoral acts, and killing by misadven-

ture (accidental, nonnegligent killing while engaged in a lawful act). These

excusing conditions establish that we cannot prejudge an action as wrong

merely because it is a killing.6 I hereafter assume this morally neutral sense of

“kill” to see where it can and should take us in contexts of physician-hastened

death.

Even though medical tradition has condemned physician killing, it is concep-

tually and morally open to physicians (and society) to reverse tradition and come

to the conclusion that medicine and the social context have changed and that it is

time to permit certain forms of assisted death including some forms that involve

killing.7 Medical morality has never been self-justifying, and traditional practices

and standards in medicine may, in the face of social change, turn out to be

indefensible limits on the liberty to choose. Even if, in medicine, killing is usually

wrong and letting die only rarely wrong, this outcome is contingent on the

features of the cases that typically appear in medicine. The general wrongness of

killing and the general rightness of letting die are not surprising features of the

moral world inasmuch as killings are rarely authorized by appropriate parties and

cases of letting die generally are validly authorized. Be that as it may, the frequency

with which one kind of act is justified, by contrast to the other kind of act, is not

relevant to the moral (or legal) justification of either kind of act.

The justifiability of any particular type or instance of killing is therefore an open

question, and we cannot assert without looking at a particular case (or type of

case) that killing is morally worse than allowing to die. The way to decide whether

killing is wrong in the medical circumstance of a request for hastened death is to

determine what makes it wrong in general. Causing a person’s death is wrong,

when it is wrong, not simply because someone is the responsible (or proximate)


cause but because an unjustified harm or loss to the deceased has occurred—for

example, the person has been deprived of opportunities and goods that life would

otherwise have afforded. A person must unjustifiably suffer a setback to personal

interests (a harm) that the person would not otherwise have experienced.8

It is a complicated question whether patient-authorized killing by physicians

involves any form of harm to the patient, but we can here circumvent this

question. It is a generally accepted principle even in Hippocratic medicine that

physicians may, under various conditions, legitimately harm patients to avoid

graver or more burdensome harms. Invasive surgery that requires recuperation is

a paradigm case. Because death is sometimes the more inviting of two unwelcome

outcomes, physicians may have sound reasons to help their patients by causing

the lesser harm.9 If a patient chooses death as a release from a desperate and

harmful circumstance, then killing at that patient’s request involves no clear

wrong even if it does involve a harm. Shortening life can avoid a state of burdens

to a patient that is virtually uncompensated by benefits.10

This form of aid to patients might harm society by setting back its interests, and

this harm might constitute a sufficient reason not to legalize physician-hastened

death. However, this consequence would not change the status of the act as a

legitimate form of physician assistance to a patient.

THE PLACE OF AUTONOMOUS REQUESTS

A request by a competent patient for assistance in hastening death does not have

the same moral authority and binding force as a competent refusal of technology

that would sustain life. Patients have a right to refuse and, correlatively, physicians

have an obligation to comply with the refusal; but there is no comparable right or

obligation in the case of a request. Autonomous refusals by patients compel

physician nonintervention, but autonomous requests by patients do not necessa-

rily compel physician intervention. Nonetheless, it does not follow that requests

from patients fail to justify acts of physician assistance in hastened death.

In many circumstances in medicine, a request by patients for aid authorizes

assistance by a physician. For example, a request for help in reducing pain

warrants interventions to meet the request. Why is a favorable response by a

physician to a request for assistance in facilitating death by hastening it different

from a favorable response to requests for assistance in facilitating death by easing

it? The two acts of physician assistance appear to be morally equivalent as long as

there are no other differences in the cases. That is, if the disease is relevantly

similar, the request by the patient is relevantly similar, the desperateness of the

patient’s circumstance is relevantly similar, and so on, then responding to a

request to provide the means to hasten death (thought of by some as killing) is


morally equivalent to responding to a request to ease death by withdrawing

treatment, sedating to coma, and the like (thought of by many as letting die).

In cases of requested active assistance, requested passive nonintervention, and

various borderline cases between the active and the passive (e.g., administering

fatal medication when the intended goal is to ease pain or to terminally sedate), a

patient is seeking the best means to quit a life of unrelieved misery. In all such

cases, persons reach a judgment that lingering in life is, on balance, worse than

death. The person who hastens death by ingesting fatal medication, the person

who forgoes nutrition and hydration, the person who requests terminal sedation,

and the person who forgoes life-sustaining technologies may each be selecting

what for him or her is the best means to the end of unrelievable burdens.

Denial of help to a patient in these bleak circumstances renders life more

burdensome from the patient’s perspective—or, in the case of not prescribing a

fatal medication, takes the decision away from the patient. It is not important that

these acts are of a certain type (suicide, say) but only that they are acts of a type

that can be justified.

The issue is not whether valid requests by patients place a duty on physicians to

hasten death.11 The question is whether valid requests render it permissible for a

physician to lend aid in hastening death. A physician with a professional commit-

ment to help a patient die as the patient chooses has made a moral commitment

that differs from the commitment made by a physician who draws the line in

opposition to all forms of assistance in hastening death. There are intermediate

forms of physician commitment, as well. A physician who, in principle, accepts

the permissibility of assistance may still refuse to honor a particular request by a

patient for assistance in a particular case on grounds that the patient’s condition

has not reached a point beyond which standard measures of palliative support are

adequate to relieve the patient’s suffering or distress. Even a sympathetic physi-

cian who is willing to assist a patient at a specific point in the evolution of his or

her condition may justifiably refuse a patient’s premature request for a hastened

death.

IS THE DISTINCTION BETWEEN KILLING AND LETTING DIE RELEVANT

TO PHYSICIAN-ASSISTED DEATH?

Many assume that the distinction between killing and letting die is relevant to

contemporary issues of physician-assisted death. This assumption is suspect.

Some forms of physician-assisted hastening of death do not involve either killing

or letting die. A physician who prescribes a lethal medication at a patient’s request

is not the cause (the proximate cause) of the patient’s death and so does not kill

the patient. Nor does this physician let the patient die. Neither the condition of


killing nor that of letting die is satisfied. Since the prescription of fatal medication

dominates much of the current discussion about hastened death, the irrelevance

to these actions of the distinction between killing and letting die merits more

discussion than it has received.

The fundamental ethical issue about physician-assisted hastening of death is

whether it is morally acceptable for physicians to help seriously ill, injured, or

suffering persons reach their goals in a manner that both the patient and the

physician find appropriate. Of course, a decision to hasten death is not justified

merely because a patient and a physician believe that it is justified. Moral

justification of acts of hastening death requires more than patient requests and

agreements between patients and their physicians. Acts of hastening death lack

justification if a physician has inattentively misdiagnosed the case, if the patient’s

capacity to make autonomous judgments is significantly impaired, if manipula-

tive family pressures are profoundly influencing either the physician or the

patient, and the like.

If no such invalidating condition exists and the physician and the patient both

act autonomously, then the choice of a hastened death is justified. One way to

express this thesis is that unless a valid liberty-limiting principle warrants inter-

vention to prevent an act of hastened death, there is no moral basis for con-

demning or punishing the envisioned autonomous acts of patients and their

willing physicians. The question of the morality of physician-assisted hastening

of death is fundamentally the question of which, if any, liberty-limiting principle

justifiably prevents such choices from being effected. Although I have not con-

sidered whether there is any such valid liberty-limiting principle, I would be

pleased if I had prepared the way for this question to become the central

issue—freeing us from the burden of the unproductive and overvalued distinction

between killing and letting die.

Elsewhere in these Collected Essays I further discuss valid liberty-limiting

principles, with a brief reference to problems of physician-hastened death. See

the essay entitled “The Concept of Paternalism in Biomedical Ethics.”

Notes

1. The line between killing and letting die is relied on in the codes and guidelines of

numerous professional associations; see AmericanMedical Association, Council on Ethical

and Judicial Affairs, “Physician-Assisted Suicide,” Report 59, December 1993. A version of

this report was published as “Physician-Assisted Suicide,” Journal of the American Medical

Association 267 (1992): 2229–2233. See also “Physician-Assisted Suicide,” Current Opinions,

H-270.965; “Euthanasia,” Current Opinions, E-2.21; “Decisions Near the End of Life,”

Current Opinions, H-140.966; “Voluntary Active Euthanasia,” Current Opinions, H-140.987

(updated June 1996); Canadian Medical Association, Code of Ethics, secs. 18–20 (approved


October 15, 1996); Canadian Medical Association, Policy Statement, “Euthanasia and

Assisted Suicide” (June 19, 1998), which replaces the previous policy entitled “Physician-

Assisted Death” (1995); World Medical Association, Resolution on Euthanasia

(Washington, DC: World Medical Association, 2002), reaffirming the Declaration of

Euthanasia (Madrid: World Medical Association, 1987).

In the formulations of these professional bodies, as in most nations and, effectively, in all

traditions of medical ethics, killing is prohibited and letting die permitted under specified

conditions. Despite this remarkable convergence of opinion, a cogent and pertinent

analysis of the distinction between killing and letting die relevant to medicine remains

elusive; see D. Orentlicher, “The Alleged Distinction Between Euthanasia and the

Withdrawal of Life-Sustaining Treatment: Conceptually Incoherent and Impossible to

Maintain,” University of Illinois Law Review 3 (1998): 837–859; T.L. Beauchamp, ed.,

Intending Death (Upper Saddle River, NJ: Prentice Hall, 1996); J. McMahan, “Killing,

Letting Die, and Withdrawing Aid,” Ethics 103 (1993): 250–279; L.O. Gostin, “Drawing a

Line Between Killing and Letting Die: The Law, and Law Reform, on Medically Assisted

Dying,” Journal of Law, Medicine and Ethics 21 (1993): 94–101; H.M. Malm, “Killing, Letting

Die, and Simple Conflicts,” Philosophy and Public Affairs 18 (1989): 238–258; B. Steinbock

and A. Norcross, eds., Killing and Letting Die, 2nd ed. (New York: Fordham University

Press, 1994).

2. F. Cohn and J. Lynn, “Vulnerable People: Practical Rejoinders to Claims in Favor of

Assisted Suicide,” in The Case against Assisted Suicide: For the Right to End-of-Life Care,

edited by K. Foley and H. Hendin (Baltimore, MD: Johns Hopkins University Press, 2002),

230–260, esp. 246–247; D. Callahan, The Troubled Dream of Life (New York: Simon and

Schuster, 1993), chap. 2; and various articles in By No Extraordinary Means, edited by

J. Lynn (Bloomington: Indiana University Press, 1986), 227–266.

3. W. Gaylin, L.R. Kass, E.D. Pellegrino, and M. Siegler, “Doctors Must Not Kill,”

Journal of the American Medical Association 259 (1988): 2139–2140 (emphasis added). See

also L.R. Kass, “Neither for Love nor Money: Why Doctors Must Not Kill,” Public Interest

94 (1989): 25–46. It is sometimes added, as a condition, that patients must be dependent

upon life-support systems; see R.E. Cranford, “The Physician’s Role in Killing and the

Intentional Withdrawal of Treatment,” in Intending Death, edited by T.L. Beauchamp

(Upper Saddle River, NJ: Prentice Hall, 1996), 150–162, esp. 160.

4. The facts about this case have been drawn from several articles that appeared from

1996 to 1999 in Canadian newspapers and medical journals. Some parts of the case rely on

reports provided to me by Canadian physicians who investigated the case.

5. In effect, this proposal is made in J.R. Rachels, “Active and Passive Euthanasia,” New

England Journal of Medicine 292 (1975): 78–80; and D.W. Brock, “Voluntary Active

Euthanasia,” Hastings Center Report 22, no. 2 (1992): 10–22.

6. Compare Rachels, “Active and Passive Euthanasia”; J.R. Rachels, “Killing, Letting

Die, and the Value of Life,” in Can Ethics Provide Answers? And Other Essays in Moral

Philosophy, by J.R. Rachels (Lanham, MD: Rowman and Littlefield, 1997), 69–79; R.W.

Perrett, “Killing, Letting Die and the Bare Difference Argument,” Bioethics 10 (1996):

131–139; Brock, “Voluntary Active Euthanasia.”


7. Compare H. Brody and F.G. Miller, “The Internal Morality of Medicine: Explication

and Application to Managed Care,” Journal of Medicine and Philosophy 23 (1998): 384–410,

esp. 397; G. Seay, “Do Physicians Have an Inviolable Duty Not to Kill?” Journal of Medicine

and Philosophy 26 (2001): 75–91.

8. Compare A. Buchanan, “Intending Death: The Structure of the Problem and

Proposed Solutions,” in Intending Death, edited by T.L. Beauchamp (Upper Saddle

River, NJ: Prentice Hall, 1996), 34–38; M. Hanser, “Why Are Killing and Letting Die

Wrong?” Philosophy and Public Affairs 24 (1995): 175–201; and reflections on the roles of

intention and the “right not to be killed” in D.W. Brock, “A Critique of Three Objections to

Physician-Assisted Suicide,” Ethics 109 (1999): 519–547, esp. 537.

9. F.M. Kamm, “Physician-Assisted Suicide, Euthanasia, and Intending Death,” in

M.P. Battin, R. Rhodes, and A. Silvers, eds., Physician-Assisted Suicide: Expanding the

Debate (New York: Routledge, 1998), 26–49; T. Nagel, “Death,” in Mortal Questions,

by T. Nagel (Cambridge: Cambridge University Press, 1979); and F.M. Kamm, Morality,

Mortality, vol. 1 (New York: Oxford University Press, 1993), chap. 1

10. See Kamm, “Physician-Assisted Suicide, Euthanasia, and Intending Death,” and

some later formulations in her “Physician-Assisted Suicide, the Doctrine of Double

Effect, and the Ground of Value,” Ethics 109 (1999): 586–605, esp. 588ff.

11. Although I am not here raising this question, Kamm has presented an “argument for

a duty of a physician”; see Kamm, “Physician-Assisted Suicide, the Doctrine of Double

Effect, and the Ground of Value,” 589.


8

THE EXPLOITATION OF THE ECONOMICALLY

DISADVANTAGED IN PHARMACEUTICAL

RESEARCH

Many media reports, books, articles, and government documents serve up searing

criticisms of the power and influence of the pharmaceutical industry. The

industry as a whole stands accused of a sea of injustices and corruptions, including

aggressive and deceptive marketing schemes, exploitative uses of research sub-

jects, a corrupting influence on universities, a shameful use of lobbying, suppres-

sion of vital data, bias and amateurism in the presentation of data, conflicts of

interest that bias research investigators, and corruption of the clinical judgment of

medical students and practicing physicians.1

Each of these charges derives from concern about some form of unjust influence

exerted by pharmaceutical companies. I cannot here consider the full array of

alleged forms of influence. I telescope to one area: the recruitment and enrollment

in clinical research of vulnerable human subjects, in particular the economically

disadvantaged. I focus on the charge that subjects in clinical trials are exploited by

manipulative and unfair payment schemes. I treat three problems. The first is the

problem of whether the economically disadvantaged constitute a vulnerable

group. I argue that classification as a “group” is a misleading characterization

that may cause paternalistic overprotection. The second problem is whether the

vulnerable poor are exploited by payments that constitute either an undue

influence or an undue industry profit. I argue that such assessments should be

made situationally, not categorically. The third problem is whether the poor give

“compromised” or nonvoluntary consents. I argue that this third problem, like

the second, is nuanced but practically manageable, and therefore that pharma-

ceutical research involving the poor and vulnerable can be carried out in an

ethically responsible manner. Whether the research is so conducted is another

matter—an empirical problem beyond the scope of my argument.

Some writers seem to argue that these problems are pseudo-problems in

developed countries because the research oversight system for clinical research

protects subjects against excessive risk and exploitation while demanding good

science and clinical promise. If the system fails, this argument goes, then the

system should be repaired, thereby resolving concerns about exploitation.2 This

approach is understandable and tempting, but overly sanguine. There is no

evidence that research oversight systems have a firm grasp on problems of

exploitation of the economically disadvantaged, let alone that these problems

can safely be dismissed. Moreover, I intend my argument to apply globally, in

all contexts of pharmaceutical research, irrespective of a country’s system of

oversight. Accordingly, I assume nothing about the adequacy of oversight

systems.

VULNERABILITY AND ECONOMIC DISADVANTAGE

Some commentators have maintained that we should not use humans as

research subjects under any circumstances; others have said that human

research subjects should be volunteers only, not persons induced by rewards

to participate. I will assume that neither view is defensible and that, under

specifiable conditions, it is permissible to involve humans as research subjects

and to pay them for their time and inconvenience. However, I do not presume

anything about the involvement of persons who are particularly vulnerable to

abuse or exploitation, or anything about levels of acceptable risk. Special

protections may be needed for some subjects, and sometimes it is inappropriate

to use such subjects at all. There may also be a need for strict control of the level

of risk allowed.

I will here consider only economically disadvantaged, decisionally compe-

tent persons. Several issues arise: Are these subjects vulnerable to morally

objectionable enticement into pharmaceutical research? If so, what renders

them vulnerable? Do they lack a mental capacity of resistibility to influence?

Is the worry that their inability to gain access to health care forces them to

accept research? Is it that offers of money or health care are made that are too

good to be refused? Is there something about the level of risk of harm that

renders subjects vulnerable? Answers to these questions are understandably

controversial.

E C O N OM I C A L L Y D I S A D V A N T A G E D I N P H A R M A C E U T I C A L R E S E A R C H 133

Vulnerabilities and Vulnerable Groups

Everyone agrees that ethical research requires impartial review to safeguard the

rights and welfare of all subjects and that some research involving vulnerable

subjects needs special scrutiny. Less clear is which populations, if any, should be

classified as vulnerable to inappropriate influences. Discussion of issues about the

vulnerability of research subjects, commonly referred to as “vulnerable groups,”

has primarily focused on embryos, fetuses, prisoners, children, psychiatric

patients, the developmentally disabled, those with dementia, and the like. Until

recently little attention has been paid to populations of persons who possess the

capacity to consent, but whose consent to participation in research might none-

theless be compromised, invalid, or unjust. Prisoners have been the paradigm

class, but the economically distressed provide a comparable example.

My concern is with persons who are impoverished, may lack significant access

to health care, may be homeless, may bemalnourished, and yet do have themental

capacity to volunteer in safety and toxicity (phase I) drug studies sponsored by or

managed by pharmaceutical companies. They possess the basic competence to

reason, deliberate, decide, and consent. Data suggest that somewhere between

50% and 100% of subjects who are healthy volunteers self-report that financial

need or financial reward is the primary motive for volunteering.3 Persons of this

description are certainly involved in some pharmaceutical research in North

America, though the extent of their use is not well understood.4 It is also known

that the poor are used in other parts of the world, sometimes in developing

countries, but the scope of their use is even less well studied and reported. I will

not distinguish between the economic conditions in diverse countries in which

research is conducted, but I will distinguish between circumstances in which it is

reasonable to expect that basic health care is available to subjects outside of

contact with research teams and circumstances in which basic health care is

fundamentally unavailable to research subjects.

It is often assumed that the economically disadvantaged are more vulnerable to

exploitation by pharmaceutical companies than subjects with stable economic

resources.5 This claim is intuitively attractive, but no data specific to pharmaceu-

tical research support it, and some data in nonpharmaceutical research cast doubt

on it.6 Nonetheless, I will assume its plausibility. What cannot be assumed is that

there is a clear connection between economically disadvantaged groups and

vulnerability or between vulnerability and exploitation by pharmaceutical

companies.

The class of the economically disadvantaged who are vulnerable has sometimes

been treated in the literature as narrow, at other times as broad. Those so classified

may or may not include individuals living on the streets, low-income persons who

are the sole financial support of a large family, persons desperately lacking access


to health care and physicians, and persons whose income falls below a certain

threshold level. These individuals could have been in their current situation for

many years, but they could also be temporarily unemployed, enduring a family

disruption, working through bankruptcy, or fighting a short-term illness.

The notion of a “vulnerable group” has lost much of its historical significance

in bioethics and health policy since it was first established in the 1970s, because so

many groups have been declared vulnerable—from the infirm elderly, to the

undereducated, to those with inadequate resources, to pregnant women, to

whole countries whose members lack rights or are subject to exploitation.

The language of “vulnerable groups” suggests that all members of a vulnerable

group—all prisoners, all poor people, all pregnant women, and the like—are by

category vulnerable, and perhaps incompetent to make their own decisions. The

problem is that for many groups a label covering all members of the group

overprotects, stereotypes, or even disqualifies members who are capable of

making their own decisions.7 “Vulnerable" is an inappropriate label for any

class of persons when some members of the class are not vulnerable. For example,

pregnant women as a class are not vulnerable, though some pregnant women are.

Accordingly, I will not speak of the economically disadvantaged—“the poor”—as

a categorically vulnerable group, though they have often been so categorized.8

However, I will speak of vulnerabilities. As necessary, I will focus on the forms,

conditions, and properties of vulnerability. Ideally, research ethics will some day

supply a schema of forms and conditions of vulnerability, rather than a list of

vulnerable groups. These forms and conditions would be composed of the

properties and circumstances that render a person vulnerable as a research sub-

ject—for example, incompetence to consent, misunderstanding the objectives of

research, increased risk, and socioeconomic deprivation that breaks down one’s

resistance to influence attempts.9

The Concept of Vulnerability

“Vulnerability” is poorly defined and analyzed in literature on the subject. Many

point to incapacities to give an informed consent, whereas others point to unequal

relationships of power and resources, such as those between the economically

disadvantaged and research sponsors or investigators. The notion of vulnerability

in research requires some form of influence either by a second party or by some

condition that renders the person susceptible to a harm, loss, or indignity.10

“Harm” and “loss” are here to be understood in terms of a thwarting, defeating,

or setting back of some party’s interests.11 “Harm” is sometimes constructed

broadly to include actions that cause discomfort, offense, indignity, or annoyance.

Narrower accounts view harms as setbacks to bodily andmental health.Whether a


broad or a narrow construal is preferable is not a matter I need to decide.

Everyone agrees that significant bodily harms and comparable setbacks to

significant nonbodily interests are paradigm instances of harm, and I will restrict

myself to these arenas.

The economically disadvantaged are vulnerable in several ways to influences

that introduce a significant risk of harm. Their situation leaves them lacking in

critical resources and forms of social power, especially access to economic

resources and government programs that might be brought to bear on their

behalf. They may not be able to resist or refuse acceptance of the risk involved,

requiring trade-offs among their interests.12What, then, should be done to protect

the interests of those who might be abused or exploited in pharmaceutical

research?

CATEGORICAL EXCLUSION OF THE ECONOMICALLY DISADVANTAGED

A tempting strategy is to exclude economically disadvantaged persons categori-

cally, even if they are not categorically vulnerable and even if they meet all

conditions for participation in clinical trials. This remedy would eliminate the

problem of their unjust exploitation in pharmaceutical research, but it would also

deprive subjects of the freedom to choose and would often be economically

harmful to them.

Nothing about economically disadvantaged persons justifies such an exclusion,

as a group, from participation in pharmaceutical research, just as it does not

follow from their status as disadvantaged that they should be excluded from

participation in any legal activity. There is an increased risk of taking advantage

of the economically distressed, but to exclude them altogether would be an

inexcusable form of discrimination. Exclusionary protections of competent, dis-

advantaged persons may only serve to further marginalize, deprive, stigmatize, or

discriminate against them. Many economically disadvantaged persons believe

that participation in research is a worthy and personally satisfying social con-

tribution as well as a means to the reduction of their economic plight. To them,

research provides an opportunity to earn money in exchange for short-term and

unskilled effort, whereas exclusion from participation is an unjust deprivation of

liberty and an offensive form of paternalism.

Consider the weakly analogous case of prisoners, who have long been the

paradigm of competent persons who are categorically excluded from phase

I clinical trials. The right to volunteer as a research subject is denied to prisoners

in U.S. federal policy on grounds of the potential for manipulation or coercion in

penal institutions. Were this same potential to exist for economically disadvan-

taged persons, the same categorical exclusion might be appropriate. However, this


problem needs to be examined in each context to see if persons who have the

capacity to give an informed consent are nonetheless not able to consent freely in

that context.

Assume, for the moment, both that we can obtain consents that are adequately

informed and voluntary and that the research is welcomed by subjects as a source

of income and possibly also as a source of contact with health care authorities.

Under these conditions, a subject’s situation of economic disadvantage is irrele-

vant. These subjects will bemore advantaged by research participation than would

other potential subjects, because the money earned and the access to health

authorities will be more significant to them. To exclude voluntary subjects from

participation would doubly disadvantage them, depriving them of money in their

already economically disadvantaged circumstance.

PAYMENT AS INDUCEMENT: PROBLEMS OF UNDUE INFLUENCE

AND UNDUE PROFIT

Payments and rewards such as free health care offered to potential pharmaceutical

research subjects present issues about the acceptability of inducements in

exchange for services. Problems of voluntariness are prominent.

Voluntariness and Types of Influence

A person acts voluntarily if he or she wills an action without being under the

controlling influence of another person or condition. Of course not all influences

are controlling, and not all controlling influences are unwarranted. What, then, are

we to say about payments to potential subjects? Are they controlling? Are they

warranted?

Three categories of influence need to be distinguished: persuasion, manipula-

tion, and coercion.13 In persuasion, a person believes in something through the

merit of reasons another person advances. This is the paradigm of an influence

that is not controlling and that is warranted. Manipulation, by contrast, involves

getting people to do what the manipulator wants through a means other than

coercion or persuasion. In recruitment for clinical studies, the most likely forms

of manipulation are (1) informational manipulation that nonpersuasively alters a

person’s understanding and (2) offers of rewards. Critics of pharmaceutical

company recruiting practices have accused companies of informational manip-

ulation through withholding critical information and through misleading exag-

geration. Other critics have said that offers of money and health care are

excessively attractive. Finally, coercion occurs if and only if one person either


intentionally and physically compels another or intentionally uses a credible and

severe threat of harm or force to control another.14 Some threats will coerce

virtually all persons, whereas others will coerce only a few persons. Coercion

itself is probably rare in the setting of clinical research, but a sense of being left

without any significant alternatives other than acceptance of an offer that one

would prefer to decline may not be uncommon. I turn now to that problem.

Constraining Situations

Subjects do sometimes report feeling heavily pressured to enroll in clinical trials,

even though their enrollment is classified as voluntary.15 These individuals may be

in desperate need of some form of medication or research may be a vital source of

income. Attractive offers such as free medication, in-clinic housing, and money

can leave a person with a sense of having no meaningful choice but to accept

research participation. Influences that many individuals easily resist are felt by

these potential subjects as constraining them to a particular choice.

In these constraining situations—sometimes misleadingly called coercive situa-

tions—there is no coercion, strictly speaking, because no one has physically

compelled another or intentionally issued a threat in order to gain compliance.

A person feels controlled by the constraints of a situation such as severe illness or

lack of food and shelter, rather than by the design or threat of another person.

Sometimes people unintentionally make other persons feel “threatened” by their

actions, and sometimes illness, powerlessness, and lack of resources are perceived

as threats of harm that a person feels compelled to prevent. No doubt these

situations are significant constraints on choice, though not ones that involve

compulsion or threats.

The prospect of another night on the streets or another day without food could

constrain a person to accept an offer of shelter and payment, just as such

conditions could constrain a person to accept a job cleaning up hazardous

chemicals or sewers that the person would otherwise not accept. The psycholo-

gical effect on persons forced to choose may be similar, and a person can

appropriately say in both cases, “I had no choice; it was unthinkable to refuse

the offer.”

PAYMENT, UNDUE INFLUENCE, AND UNDUE PROFIT

In constraining situations, monetary payments and related offers such as shelter

or food give rise to questions of undue influence on the one hand and undue profit

on the other. The “Common Rule” in the United States requires investigators to


“minimize the possibility of” coercion and undue inducement, but it does not

define, analyze, or explain these notions.16 Issues of exploitation, undue induce-

ment, and undue profit are inadequately handled in the bioethics literature as

well.

Monetary payments seem unproblematic if the payments are welcome offers

that persons do not want to refuse and the risks are at the level of everyday

activities. Becoming a research subject, under these conditions, seems no different

than accepting a job offer or agreeing to a contract that, without payment, one

would not accept. Indeed, the offer of research involvement is an offer of a job or

contract in which mutual benefit can and should occur (setting aside questions of

sanitary conditions, impartial committee review and approval, and the like).17

However, inducements become increasingly problematic (1) as risks are increased,

(2) as more attractive inducements are offered, and (3) as the subject’s economic

disadvantage is exacerbated. These are the three principal conditions to be

considered in assessing whether exploitation occurs.18

The heart of the problem of exploitation is that subjects are situationally

disadvantaged and without viable alternatives, feel forced or compelled to

accept attractive offers that they otherwise would not accept, and assume

increased risk in their lives. As these conditions are mitigated, the problem of

exploitation diminishes and may vanish; but as these conditions are heightened,

the potential for exploitation increases.

As this formulation suggests, the condition of an irresistibly attractive offer is a

necessary condition of “undue inducement,” but this condition is not by itself

sufficient to make an inducement undue. There must also be a risk of harm of

sufficient seriousness that the person’s welfare interest is negatively affected by

assuming it, and it must be a risk the person would not ordinarily assume.19 I will

not try to pinpoint a precise threshold of risk, but it would have to be above the

level of such common job risks as those of unskilled construction work.

Inducements are not undue, then, unless they are both above the level of standard

risk (hence “excessive” in risk) and irresistibly attractive (hence “excessive” in

payment) in light of a constraining situation. Although these offers are not

coercive, because no threat of excessive risk or of taking money away from the

person is involved, the offer can be manipulative. Indeed, since irresistibly

attractive payment is involved, these offers almost certainly should be categorized

as manipulative.

Undue inducements should be distinguished from undue profits, which occur

from a distributive injustice of too small a payment, rather than an irresistibly

attractive, large payment. In the undue profit situation, the subject gets an unfair

payment and the sponsor gets more than is justified. Often, I think, this is what

critics of pharmaceutical research believe happens: The subjects are in a weak to

nonexistent bargaining situation, are constrained by their poverty, and are given a


pitifully small amount of money and unjust share of the benefits, while companies

reap unseemly profits from the research. If this is the worry, the basic question is

how to determine a nonexploitative, fair wage.

How, then, should these two moral problems of exploitation—undue induce-

ment (unduly large and irresistible payments) and undue profit (unduly small

and unfair payments)—be handled? One possible answer is that if this research

involves exceptional, and therefore excessive, risk (i.e., risk having more than a

minor increment above minimal risk), it should be prohibited categorically for

healthy subjects, even if a good oversight system is in place. This answer is

appealing, but we would still need to determine what constitutes excessive risk,

irresistibly attractive payment, unjust underpayment, and constraining situa-

tions—all difficult problems that seem largely ignored in literature on the subject.

As if these problems are not difficult enough, there is a deeper moral dilemma

than the analysis thus far suggests. To avoid undue influence, payment schedules

must be kept reasonably low, approximating an unskilled labor wage—or possibly

even lower. Even at this low level, payment might still be sufficiently large to

constitute an undue inducement for some subjects. As payments are lowered

down a continuum of wages to avoid undue influence—that is, to avoid making

an excessively attractive offer—it is predictable that research subjects in some

circumstances will be recruited largely or perhaps entirely from the ranks of the

economically disadvantaged. Somewhere on this continuum the amount of

money paid will be so little that it is exploitative by virtue of undue profits yielded

by taking advantage of a person’s misfortune. If the payment scales were reversed

and run up the continuum—that is, increased to avoid undue profit—they would

at some point become high enough to attract persons from the middle class. At or

around this point, the offers would be declared exceptionally (and therefore

excessively) attractive, and would for this reason be judged undue inducements

for impoverished persons.20 This problem becomes a more profound problem of

potential injustice as the pool of research subjects is composed more or less

exclusively of the economically disadvantaged. (See, further, the section on

Background Justice later in this essay.)

The argument thus far yields the conclusion that exploitation occurs either

when there is undue profit, offering too little for services, or when there is undue

influence, offering too much for services. The most straightforward way to avoid

such exploitation is a golden mean that strikes a balance between a rate of

payment high enough that it does not exploit subjects by underpayment and

one low enough that it does not exploit by undue inducement. If this is right, there

is no a priori way to set a proper level of payment (e.g., at the level of unskilled

labor) because that level might not satisfy the golden mean standard. Payment at

the level of unskilled labor rates might itself be exploitative, either by creating an

undue influence or generating an undue profit. The general objective should be


that the research sponsor pay a fair wage at the golden mean for moderate-risk

studies and not increase the wage in order to entice subjects to studies with a

higher level of risk.21 If this mean is unattainable, then any such research will be

exploitative: There will be either an undue profit by underpayment or an undue

influence by overpayment.22

If this analysis is correct, there may be situations in which payments that are too

high (creating undue inducements) are, at the same time, payments that are too

low (creating undue profits). To the desperate, $.25 per hour or $10 per hour

might be irresistibly attractive while distributively unfair. Critics of the pharma-

ceutical industry often seem to suggest that something like this problem is

common in the research context and that pharmaceutical companies routinely

take advantage of these situations. If this is the charge, critics of the pharmaceu-

tical industry could turn out to be right. As far as I know, such questions have

never been carefully studied, and I have not tried to decide these questions. From

what I know, at least some contexts of research conducted in North America do

not involve either undue inducement or undue profit, but my argument here does

not require such an assumption about North America or any other part of the

world. My argument only tries to locate the moral problems and to consider

possible paths to resolution.

A conclusion that is tempting to draw at this point is that pharmaceutical

companies are morally obligated to avoid any situation in which the dilemma I

have traced over payments cannot be satisfactorily resolved by recourse to the

golden mean. That is, companies would be morally prohibited from sponsoring

research in circumstances that make it exploitative by paying too much or too

little. Critics of the industry would presumably be happy with this outcome, and

prosperous pharmaceutical companies would also be happy with it, because it

would tell themwhere not to locate their research in order to avoid criticism. They

would then locate, or relocate, elsewhere.

I am not convinced, however, that this is the best outcome. Some moderate-

risk research seems nonexploitative in adequately supervised circumstances even

if an offer is irresistibly generous and choices are constrained—and perhaps even

if the company could afford to pay the subjects more. However, supporting this

conclusion would take more argument than I have provided, and I will not pursue

it further. I make only one comment to frame the issue: An important reason for

caution about prohibiting research and encouraging pharmaceutical companies

to pull out of poor communities is that payments for studies are a vital source of

needed funds for the economically disadvantaged as well as a way to build an

infrastructure of jobs in these communities. One of the few readily available

sources of money for some economically distressed persons are jobs such as day

labor that expose them to more risk and generate less money than the payments

generated by participation in phase I clinical trials.23 To deny these persons the


right to participate in clinical research on grounds of potential exploitation can be

paternalistic and demeaning, as well as economically distressing.

Moreover, many pharmaceutical companies have the resources to make life

better for subjects not only by paying them a decent wage but also by improving

their lives in other respects. For example, in some countries research centers could

afford to include in their budgets a range of health care benefits (including

ancillary medical care not elsewhere available), financial planning, and personal

counseling. These centers could make a tangible contribution to the education

and social welfare of disadvantaged persons, not merely a contribution to their

financial welfare. These programs too must be viewed as offers of benefits (and in

this respect no different from monetary offers) and therefore as potentially

creating problems of undue influence (also like monetary offers). At the same

time, it would be unfeeling to disregard the contribution that such programs

could make to the lives of the economically disadvantaged.

A maneuver that should be prohibited, as a matter of policy, is a sliding scale

that would raise or lower payments based on an attempted assessment of risk of

injury, pain, or discomfort. This maneuver would simply magnify concerns about

exploitation. To reward taking risks by offering increased payment would too

easily induce the most disadvantaged to take the highest risks; and to take the view

that the higher the risk, the less one should pay in order to avoid the problem of

undue inducement would be to discriminate by penalizing persons for their

economic disadvantage. These are morally unacceptable outcomes that should

be avoidable by not conducting research with increased risk in any location in

which this problem might arise.

COMPROMISED CONSENT

Compromised consent presents another concern about the vulnerability of

research subjects. The typical concerns about consent in pharmaceutical research

center either on whether the consent is adequately informed or on whether

subjects have the ability to give an informed consent to research participation. I

am not concerned with the ability to consent, as I am discussing only subjects who

do have this ability. I am concerned only with the process used in the research

context to obtain consent and the ways in which consent can be compromised.24

The most frequently mentioned problems of consent in pharmaceutical

research are undisclosed data, unknown risks, and inadequate discussion with

trial participants about risks and inconveniences.25 Some practices of obtaining

consent are clearly shams, where consent is invalid even if a written consent

document exists. Consider a paradigm case of this problem: On November 2,

2005, Bloomberg News reported that SFBC International, Inc., a global clinical


research and drug development services company, had recruited a number of

economically disadvantaged participants for drug-testing contracts by making

inadequate disclosures regarding the risk of the research. SFBC was reported to

have manipulated participants through increases of compensation during the

course of the research and to have used illegal, non–English-speaking immigrants

fearful of exportation, with some used in multiple trials. The amount of compen-

sation was reported in some cases to be very high (an undue influence) and in

other cases very low (an undue profit).26 If these reports are even half correct,

SFBC engaged in a blatant series of violations of nondisclosure, in the manipula-

tion of consent, and possibly even in coercion. This case is without moral subtle-

ties, and all morally sensitive persons would recognize it as an egregious wrong.

However, there are many subtle and difficult problems of consent, including

ways in which information is presented initially and then throughout the trial.27

Without appropriate monitoring of consent, even an initially informed and valid

consent can become uninformed and invalid. The appropriate framework for

obtaining and updating consent in pharmaceutical research is not a single event of

oral or written consent, but a supervised, multistaged arrangement of disclosure,

dialogue, and permission giving that takes place after adequate oversight of a

protocol.

One stage in the informational process should occur during the interview that

is scheduled when a potential subject responds to an advertisement announcing

an upcoming study. Potential subjects should learn basic facts such as whether the

study is an inpatient or outpatient study, the study’s duration, forms of testing,

disqualifying medical conditions, requirements such as age and smoking status,

and how eating and drinking are controlled during the study. The goals should be

to inform people of what they will experience and to prevent people from coming

to the clinic or research center who will be excluded from the studies.

A second stage in the informed consent process should be composed of

discussions that occur during in-house screening and laboratory testing. If

laboratory values and other inclusion criteria are met, potential subjects should

be scheduled for an interview and physical examination. If potential subjects

come back, the nature and content of the consent form should be explained.

At this stage potential subjects should be told about the drug being tested, the

route of administration, diseases for which the drug is being developed, the

amount of blood sampling, unusual risks, and the like. They should be asked

whether they have understood why they are being screened, and there should be

an established way of testing for adequate understanding.

A third stage should be a session devoted to the formalities of informed

consent: disclosure, discussion to increase understanding, and the like. These

sessions should be limited to individuals or small groups, with informed profes-

sionals in the room. The goal should be a comprehensible explanation of the study


and the authorization form. Potential volunteers should come to know enough to

elect to participate in the study or decide not to participate.

Answering questions, eliciting the concerns and interests of the subject, and

establishing a climate that encourages questions are necessary conditions of the

consent process, which is essential to justified human-subjects research. To fail to

conform research to the conditions in the three stages mentioned in this section is

to obtain what I am calling a compromised consent. Research that does not meet

these conditions involves an inexcusable disregard of the rights of subjects, no

matter who conducts it or where it occurs.

CONSIDERATIONS OF BACKGROUND JUSTICE

Several issues I have been considering should also be set against a richer back-

ground of concerns about social justice. Many considerations of justice suggest

that society should take a greater interest in the economically disadvantaged than

it often does—not because they have become or might become research subjects,

but because of their status as exploitable members of society who are poorly

served by social services.

One problem of pharmaceutical research is the potential injustice that occurs

when a benefit is generated for the well-off through a contribution to research

made largely by disadvantaged members of society. Research, as currently con-

ducted, places aminority of persons at risk in order that others, or sometimes only

the well-off, benefit. However, how the burdens of research involvement (if they

are burdens, which is an underanalyzed idea) should be distributed in a popula-

tion has never been established in any authoritative code, document, policy, or

moral theory. The notion of “equity,” commonly used in current discussions, is

often poorly analyzed and poorly implemented. Today, it remains unclear, and

controversial, when a research endeavor becomes a morally objectionable or

marginal form of co-opting the economically disadvantaged for the benefit of

the privileged.

Suppose, however, that a pharmaceutical company were to target the econom-

ically disadvantaged in its recruitment of subjects and were to use exclusively these

subjects in its research. This activity raises some seldom discussed questions about

disproportionate uses of the economically disadvantaged. It may be appropriate

to have selection criteria that demand that the research enterprise not specifically

target the economically disadvantaged and that the percentages of economically

disadvantaged subjects in research protocols be restricted, while ensuring that

research subjects are drawn from suitably diverse populations. This standard does

not imply that it is always unfair to recruit (even heavily) from an economically

disadvantaged population, only that recruitment introduces problems of


distributive justice not covered by the issues considered previously. If this judg-

ment is right, then pharmaceutical companies should have policies requiring that

they carefully monitor the numbers and percentages of economically disadvan-

taged persons.

One issue is whether there is an upper level or quota of acceptable participation

by members of the economically disadvantaged in a protocol or research center.

Setting a quota would disqualify some potential subjects, which has the potential

of being unfair to them. If no categorical problem exists over the involvement of

economically disadvantaged adults as subjects, as I suggested earlier, then why

should we worry about numbers, quotas, or percentages? The relevant conditions

would seem to be that each individual in a trial, whether economically disadvan-

taged or not, is physically qualified, does not have a disqualifying condition, gives

an informed consent, and the like. If a given individual qualifies, there seems no

reason to set a numerical limit on the number of studies in which he or she can

participate.

This approach should be taken seriously, but it does not adequately address the

background considerations of justice that should be attended to. Requiring

diversity in a subject pool potentially protects against several conditions that

might otherwise too easily be tolerated in the research process—for example,

that drug studies would be conducted in dreary and depersonalizing environ-

ments, that the same subjects would be used repeatedly in studies, and that a

dependency on money would set in, making subjects dependent on drug studies

for their livelihood. Related problems concern whether it is problematic for

individuals to be involved by volunteering for multiple studies, a problem that

would be exacerbated as financial inducements are increased.28

CONCLUSION

Everyone agrees with the abstract rule that we should not exploit the economically

disadvantaged in a pharmaceutical research project. I have argued, however, that

it is not easy to determine what counts as exploitation, the conditions under

which it occurs, who is vulnerable to exploitation, how to avoid it, and whether

certain trade-offs or compromises are acceptable. The recommendations I have

made about the conduct of studies involving the economically disadvantaged

have centered on vulnerabilities and exploitation. I have argued that while

pharmaceutical research need not be exploitative or unjustified when conducted

with the economically disadvantaged, these problems are more nuanced and

difficult than they have generally been treated in literature on the subject. We

also know relatively little about the conditions under which inducements are

accepted and about whether those who accept those inducements view them as


unwelcome exploitations or as welcome opportunities.29We need to get clearer—

conceptually, morally, and empirically—about the populations involved in

research and about the difference between the unethical and the mutually bene-

ficial when inducements are offered to those in these populations. We confront a

considerable task.

Notes

1. Marcia Angell, The Truth About the Drug Companies: How They Deceive Us and What

to Do About It (New York: RandomHouse, 2005); House of Commons Health Committee,

The Influence of the Pharmaceutical Industry, Fourth Report of Session 2004-05 (London:

Stationery Office, 22 March 2005), esp. 43ff; T. S. Faunce and G. F. Tomossy, “The UK

House of Commons Report on the Influence of the Pharmaceutical Industry: Lessons for

Equitable Access to Medicines in Australia,” Monash Bioethics Review 23, No. 4: 38–42;

Department of Health and Human Services, Office of the Inspector General, Prescription

Drug Promotion Involving Payments and Gifts: Physicians’ Perspectives OEI-01-90-00481

(Washington, DC: DHHS, 1991) and also DHHS, “Pharmaceutical Company Gifts and

Payments to Providers,” inWork Plan for Fiscal Plan 2002 (Washington, DC: DHHS, 2001);

Jerome P. Kassirer,On the Take: HowMedicine’s Complicity with Big Business Can Endanger

Your Health (Oxford/New York: Oxford University Press, 2005); Dana Katz, Arthur L.

Caplan, and Jon F.Merz, “All Gifts Large and Small,”American Journal of Bioethics 3 (2003):

39–46; Merrill Goozner, The $800 Million Pill: The Truth Behind the Cost of New Drugs

(Berkeley: University of California Press, 2004); Jerry Avorn, Powerful Medicines: The

Benefits, Risks, and Costs of Prescription Drugs (New York: Vintage Books, 2005); Jerry

Avorn, M. Chen, and R. Hartley, “Scientific versus Commercial Sources of Influence on the

Prescribing Behavior of Physicians,” American Journal of Medicine 73 (1982): 4–8; Fran

Hawthorne, Inside the FDA: The Business and Politics Behind the Drugs We Take and the

Food We Eat (Hoboken, NJ: John Wiley & Sons, 2005); Jacky Law, Big Pharma: How

Modern Medicine is Damaging Your Health and What You Can Do About It (New York:

Carroll & Graf, 2006); Jennifer Washburn, University Inc.: The Corporate Corruption of

Higher Education (New York: Basic Books, 2005); A. Wazana, “Physicians and the

Pharmaceutical Industry: Is a Gift Ever Just a Gift?” Journal of the American Medical

Association 283 (2000): 373–380; Arnold S. Relman, “Separating Continuing Medical

Education from Pharmaceutical Marketing,” Journal of the American Medical Association

285 (2001): 2009–2012.

2. Ezekiel Emanuel, “Ending Concerns about Undue Inducement,” Journal of Law,

Medicine, & Ethics 32 (2004): 100–105.

3. Carl Tishler and Suzanne Bartholomae, “The Recruitment of Normal Healthy

Volunteers,” Journal of Clinical Pharmacology 42 (2002): 365–375.

4. Tom L. Beauchamp, Bruce Jennings, Eleanor Kinney, and Robert Levine,

“Pharmaceutical Research Involving the Homeless,” Journal of Medicine and Philosophy

(2002), 547–564; M. H. Kottow, “The Vulnerable and the Susceptible,” Bioethics 17 (2003):

460–471; Toby L. Schonfeld, Joseph S. Brown, Meaghann Weniger, and Bruce Gordon,


“Research Involving the Homeless,” IRB 25 (Sept./Oct. 2003): 17–20; C. E. van Gelderen,

T. J. Savelkoul, W. van Dokkum, J. Meulenbelt, “Motives and Perceptions of Healthy

Volunteers Who Participate in Experiments,” European Journal of Clinical Pharmacology 45

(1993): 15–21.

5. Glenn McGee, “Subject to Payment?” Journal of the American Medical Association 278

(July 16, 1997): 199–200; Leonardo D. de Castro, “Exploitation in the Use of Human

Subjects for Medical Experimentation: A Re-Examination of Basic Issues,” Bioethics 9

(1995): 259–268; Laurie Cohen, “To Screen New Drugs For Safety, Lilly Pays Homeless

Alcoholics,” Wall Street Journal, Nov. 14, 1996, A1, A10.

6. Scott Halpern, Jason Karlawish, David Casarett, Jesse Berlin, and David A. Asch,

“Empirical Assessment ofWhether Moderate Payments Are Undue or Unjust Inducements

for Participation in Clinical Trials,” Archives of Internal Medicine 164 (2004): 801–803.

7. These problems, including the vast range of groups and populations declared to be

vulnerable in the bioethics and public policy literature, are assembled and discussed in

Carol Levine, Ruth Faden, Christine Grady, Dale Hammerschmidt, Lisa Eckenwiler,

Jeremy Sugarman (for the Consortium to Examine Clinical Research Ethics), “The

Limitations of ‘Vulnerability’ as a Protection for Human Research Participants,”

American Journal of Bioethics 4 (2004): 44–49; Debra A. DeBruin, “Reflections on

Vulnerability,” Bioethics Examiner 5 (2001): 1, 4, 7. For examples of treatments of these

issues that assume broadly stated—and numerous classes—of vulnerable groups, see

Council for International Organization of Medical Sciences, International Ethical

Guidelines for Biomedical Research Involving Human Subjects, http://www.cioms.ch/fra-

me_guidelines; Ruth Macklin, “Bioethics, Vulnerability, and Protection,” Bioethics 17

(2003): 472–486; and Laura B. Sutton, et al., “Recruiting Vulnerable Populations for

Research: Revisiting the Ethical Issues,” Journal of Professional Nursing 19 (2003): 106–112.

8. See Ruth Macklin, “‘Due’and ‘Undue’ Inducements: On Paying Money to Research

Subjects,” IRB 3 (1981): 1–6; Tishler and Bartholomae, “The Recruitment of Normal

Healthy Volunteers.”

9. See Debra A. DeBruin, “Looking Beyond the Limitations of ‘Vulnerability’:

Reforming Safeguards in Research,” The American Journal of Bioethics 4 (2004): 76–78;

National Bioethics Advisory Commission, Ethical and Policy Issues in Research Involving

Human Participants, vol. 1 (Bethesda, MD: Government Printing Office, 2001); Kenneth

Kipnis, “Vulnerability in Research Subjects: A Bioethical Taxonomy,” National Bioethics

Advisory Commission, Ethical and Policy Issues in Research Involving Human Participants,

vol. 2 (Bethesda, MD: Government Printing Office, 2002), G1–13

10. Gail E. Henderson, Arlene M. Davis, and Nancy M. P. King, “Vulnerability to

Influence: A Two-Way Street,” American Journal of Bioethics 4 (2004): 50–53

11. Joel Feinberg, Harm to Others, vol. 1 of The Moral Limits of the Criminal Law

(New York: Oxford University Press, 1984), 32–36.

12. Cf. Anita Silvers, “Historical Vulnerability and Special Scrutiny: Precautions against

Discrimination in Medical Research,” American Journal of Bioethics 4 (2004): 56–57.


(New York: Oxford University Press, 1986), Chapter 10; R. B. Cialdini, Influence:


http://www.cioms.ch/frame_guidelines

http://www.cioms.ch/frame_guidelines

The Psychology of Persuasion (New York: Quill WilliamMorrow, 1993); Patricia Greenspan,

“The Problem with Manipulation,” American Philosophical Quarterly 40 (2003): 155–164.

14. See Robert Nozick, “Coercion,” in Philosophy, Science andMethod: Essays in Honor of

Ernest Nagel, ed. Sidney Morgenbesser, Patrick Suppes, and Morton White (New York:

St. Martin’s Press, 1969), 440–472; and Bernard Gert, “Coercion and Freedom,” inCoercion:

Nomos XIV, ed. J. Roland Pennock and JohnW. Chapman (Chicago: Aldine, Atherton Inc.

1972), 36–37.

15. See Sarah E. Hewlett, “Is Consent to Participate in Research Voluntary?” Arthritis

Care and Research 9 (1996): 400–404; Hewlett, “Consent to Clinical Research—Adequately

Voluntary or Substantially Influenced?” Journal of Medical Ethics 22 (1996): 232–236; Robert

M. Nelson and Jon F. Merz, “Voluntariness of Consent for Research: An Empirical and

Conceptual Review,” Medical Care 40 (2002) Suppl., V69–80; and Nancy E. Kass, et al.,

“Trust: The Fragile Foundation of Contemporary Biomedical Research,” Hastings Center

Report 25 (September–October 1996): 25–29.

16. Common Rule for the Protection of Human Subjects, U.S. Code of Federal

Regulations, 45 CFR 46.116 (as revised Oct. 1, 2003); and Emanuel, “Ending Concerns

about Undue Inducement,” 101.

17. The justification of monetary inducement in terms of mutual benefit is defended by

Martin Wilkinson and Andrew Moore, “Inducement in Research,” Bioethics 11 (1997):

373–389; and Wilkinson and Moore, “Inducements Revisited,” Bioethics 13 (1999): 114–

130. See also Christine Grady, “Money for Research Participation: Does It Jeopardize

Informed Consent?” American Journal of Bioethics 1 (2001): 40–44.

18. In some cases, the offers may blind prospective subjects to risks and inconveniences,

but such blindness is a concern about adequately informed consent.

19. Procedures performed in the conduct of research protocols do present real risks of

injury, and a threshold level of appropriate or acceptable risk must be established. Some

have suggested a threshold standard of “minimal risk,” which is defined in federal regula-

tions: “Minimal risk means that the probability and magnitude of harm or discomfort

anticipated in the research are not greater in and of themselves, than those ordinarily

encountered in daily life or during the performance of routine physical or psychological

examinations or tests.” 45 CFR 46.102i (as revised Oct. 1, 2005). For many adult popula-

tions, a more realistic threshold standard is a “minor increase above minimal risk.”

Published evidence and commentary indicate that most phase I and phase IV studies of

drugs present such minor increases over minimal risk. For studies of risks and safety in

pharmaceutical research, see: M. Sibille et al., “Adverse Events in Phase one Studies:

A Study in 430 Healthy Volunteers,” European Journal of Clinical Pharmacology 42 (1992):

389–393; P. L. Morselli, “Are Phase I Studies without Drug Level Determination

Acceptable?” Fundamental and Clinical Pharmacology 4 Suppl. 2 (1990): 125s–133s;

M. Orme, et al., “Healthy Volunteer Studies in Great Britain: The Results of a Survey

into 12Months Activity in this Field,” British Journal of Clinical Pharmacology 27 (February

1989): 125–133; H. Bostr€om, “On the Compensation for Injured Research Subjects in

Sweden,” in President’s Commission for the Study of Ethical Problems in Medicine and

Biomedical and Behavioral Research, Compensating for Research Injuries: The Ethical and


Legal Implications of Programs to Redress Injured Subjects, Appendix, 309–322 (Washington,

DC: U.S. Government Printing Office, Stock No. 040-000-00455-6, 1982; the Appendix to

this Report is Stock No. 040-000-00456-4); P. V. Cardon, F. W. Dommel, and

R. R. Trumble, “Injuries to Research Subjects: A Survey of Investigators,” New England

Journal of Medicine 295 (1976): 650–654; C. J. D. Zarafonetis, P. A. Riley, Jr., P. W.Willis, III,

L. H. Power, J. Werbelow, L. Farhat, W. Beckwith, and B. H. Marks, “Clinically Significant

Adverse Effects in a Phase I Testing Program,” Clinical Pharmacology and Therapeutics 24

(1978): 127–132; J. D. Arnold, “Incidence of Injury During Clinical Pharmacology Research

and Indemnification of Injured Research Subjects at the Quincy Research Center,” in

President’s Commission for the Study of Ethical Problems in Medicine and Biomedical

and Behavioral Research, Compensating for Research Injuries: The Ethical and Legal

Implications of Programs to Redress Injured Subjects, Appendix, 275–302 (Washington,

DC: U.S. Government Printing Office, Stock No. 040-000-00455-6, 1982; the Appendix to

this Report is Stock No. 040-000-00456-4).

20. See Neal Dickert and Christine Grady, “What’s the Price of a Research Subject?:

Approaches to Payment for Research Participation,” New England Journal of Medicine 341

(1999): 198–203; David Resnick, “Research Participation and Financial Inducements,”

American Journal of Bioethics 1 (2001): 54–56; Wilkinson and Moore, “Inducement in

Research,” 373–389; Macklin, “‘Due’ and ‘Undue’ Inducements: On Paying Money to

Research Subjects,” 1–6; Lisa H. Newton, “Inducement, Due and Otherwise,” IRB:

A Review of Human Subjects Research 4 (March 1982): 4–6; Ruth Macklin, “Response:

Beyond Paternalism,” IRB: A Review of Human Subjects Research 4 (March 1982): 6–7.

21. Compare the rather different conditions proposed by de Castro, “Exploitation in the

Use of Human Subject for Medical Experimentation: A Re-Examination of Basic Issues,”

264, and by Beauchamp, Jennings, Kinney, and Levine, “Pharmaceutical Research

Involving the Homeless.” See also Schonfeld, Brown, Weniger, Gordon, “Research

Involving the Homeless: Arguments against Payment-in-Kind,” 17–20.

22. One issue that will constantly have to be monitored is how to keep the offer from

being irresistible. This elicits the question “irresistible by what standard?” An offer that is

irresistible to one personmay be resistible to another, or what one greets as a welcome offer,

the other may greet as unwelcome; and this is so even if the offers are identical. How an

offer is perceived and whether it will be accepted depend on the subjective responses of the

persons who receive the offer. On this analysis, undue influence in the acceptance of

increased risk occurs only if the person receiving the offer finds it irresistible.

23. See Nik Theodore, Edwin Melendez, and Ana Luz Gonzalez, “On the Corner: Day

Labor in the United States,” http://www.sscnet.ucla.edu/issr/csup/uploaded_files/

Natl_DayLabor-On_the_Corner1.pdf.

24. I am also not concerned with situations in which there is a signed and adequately

informed consent and yet the consent is invalid. For example, the protocol may be

scientifically inadequate, there may have been an unjust selection of subjects, risks and

inconveniences may not warrant the research, etc. I am not concerned with such invali-

dating conditions. I am also not here concerned with problems of consent deriving from

the problems of undue inducement (controlling influences and constraining situations)


http://www.sscnet.ucla.edu/issr/csup/uploaded_files/Natl_DayLabor-On_the_Corner1.pdf

http://www.sscnet.ucla.edu/issr/csup/uploaded_files/Natl_DayLabor-On_the_Corner1.pdf

examined previously. Constraining situations may compromise the voluntariness of

consent, but we need not revisit that problem here.

25. House of Commons Health Committee, The Influence of the Pharmaceutical

Industry, 49ff.

26. Kerry Dooley Young and David Evans, with reporting by Michael Smith in Rio De

Janeiro, “SFBC’s Top Two Officials Quit Amid U.S. Senate Probe,” Bloomberg News, Jan. 3,

2006. The original report of problems was by the same reporters, Bloomberg News, Nov. 2,

2005, “Poor Latin Immigrants SayMiami Test Center ‘Is Like a Jail’.” Bloomberg has issued

numerous later reports in the case.

27. See Christopher K. Daugherty, Donald M. Banik, Linda Janish, and Mark J. Ratain,

“Quantitative Analysis of Ethical Issues in Phase I Trials,” IRB 22 (May–June 2000): 6–14,

esp. 12–13.

28. Paul McNeill, “Paying People to Participate in Research: Why Not?” Bioethics 11

(1997): 390–396; Beauchamp, Jennings, Levine, and McKinney, “Pharmaceutical Research

Involving the Homeless”; Michael A. Grodin and Leonard H. Glantz, eds. Children as

Research Subjects: Science, Ethics, and Law (New York: Oxford University Press; 1994),

193–214; National Commission for the Protection of Human Subjects of Biomedical and

Behavioral Research, The Belmont Report: Ethical Guidelines for the Protection of Human

Subjects (Washington, DC: DHEW Publication (OS) 78-0012, 1978).

29. See David Casarett, Jason Karlawish, and David A. Asch, “Paying Hypertension

Research Subjects: Fair Compensation or Undue Inducement,” Journal of General Internal

Medicine 17 (2002): 651–653; and Scott Halpern, Jason Karlawish, David Casarett, Jesse

Berlin, and David A. Asch, “Empirical Assessment of Whether Moderate Payments Are

Undue or Unjust Inducements for Participation in Clinical Trials,” 801–803.


Part III

THEORY AND METHOD

9

PRINCIPLES AND OTHER EMERGING

PARADIGMS IN BIOETHICS

If a history of recent biomedical ethics were to be written, it would encompass

several disciplines, including the health professions, law, biology, the social and

behavioral sciences, theology, and philosophy. Principles that could be understood

with relative ease by the members of these disciplines figured prominently in the

development of biomedical ethics during the 1970s and early 1980s. Principles were

used primarily to present frameworks of evaluative assumptions or general prem-

ises underlying positions and conclusions.1 However, beginning in the mid-1980s,

the paradigm of a system of principles was aggressively challenged. Several alter-

natives have since been proposed, including revivals of casuistry and virtue theory.

These developments should be welcomed in bioethics, because they have improved

the range, precision, and quality of thought in the field. However, the various

proposed alternative approaches do not replace principles. The leading alternatives

are thoroughly compatible with a paradigm of principles. Indeed, these seemingly

different frameworks of norms are mutually supportive.

I will begin my argument to this conclusion by outlining the nature of a principle-

based approach to ethics, concentrating on the book James Childress and I wrote in

the mid-1970s entitled Principles of Biomedical Ethics (4th edition, 1994). After

sketching our ethical framework, I will point to some limitations of the model and

indicate how those limitations should be handled in the first section, Principles as a

Starting Point. I will consider the nature and limits of three proposed alternatives to a

principle-based approach in the second section, Alternative Paradigms.

PRINCIPLES AS A STARTING POINT

Principle-based ethical theories emphasize impartial moral obligations, but

“principles” should not be defined in terms of obligations. Moral principles are

simply relatively general norms of conduct that describe obligations, permissible

actions, and ideals of action. A principle is a regulative guideline stating condi-

tions of the permissibility, obligatoriness, rightness, or aspirational quality of an

action falling within the scope of the principle. If principles are adequately

expressed, more particular or specific moral rules and judgments are supported

by, though not deduced from, the principles. For example, principles of justice

provide support for particular rules and judgments regarding equal treatment,

fair taxation, and just compensation.

A Principle-Based Paradigm

The paradigm that Childress and I defend is that various principles worthy of

acceptance in bioethics can be grouped under four general categories: (1) respect

for autonomy (a principle of respect for the decision-making capacities of auton-

omous persons), (2) nonmaleficence (a principle of avoiding the causation of

harm to others), (3) beneficence (a group of principles for providing benefits and

balancing benefits against risks and costs), and (4) justice (a group of principles

for distributing benefits, risks, and costs fairly).

We do not sharply distinguish between rules and principles in our analyses.

Both are general action guides. In addition to substantive rules of truth telling,

confidentiality, privacy, fidelity, and the like, there are authority rules regarding

who may and should perform actions, including rules of surrogate authority,

rules of professional authority, and rules of distributional authority that deter-

mine who should make decisions about the allocation of scarce medical

resources. Procedural rules are also important in bioethics, because they establish

procedures to be followed, such as procedures for determining eligibility for

scarce medical resources.

Principles, being more abstract than rules, leave considerable room for

judgment about individual cases and policies. However, principles should be

conceived neither as so weak that they are mere rules of thumb nor as so

strong that they assert absolute requirements. Insofar as they assert obligations

(but not insofar as they are used to frame ideals), principles are firm obliga-

tions that can be set aside only if they come into conflict with and do not

override another obligation. In cases of a conflict of obligations, one obliga-

tion then has the potential to release the person from the other obligation.

Often some balance between two or more norms must be found that requires

154 T H E O R Y A N D M E T H O D

some part of each obligation to be discharged, but in many cases one simply

overrides the other.

This overriding of one obligation by another may seem precariously flexible, as

if moral guidelines in the end lack backbone and can be magically waived away as

not real obligations. But in ethics, as in all disciplines that confront principled

conflicts, such as law, there is no escape from the exercise of judgment in resolving

the conflicts. One function of principles is to keep judgments principled without

removing agent discretion. As long as an agent does not stray beyond the demands

of principles, it cannot be said that judgments are arbitrary or unprincipled, even

if one principle overrides another.

I do not say that every judgment intended to resolve a principled conflict must

itself be resolved by the principles that are in conflict, a manifestly false thesis.

Skillful use of principles requires judgment, which in turn depends on character,

moral insight, and a sense of personal responsibility and integrity; and these

properties of persons are neither principles nor merely a way of conforming to

principle. Sensitive, prudent, or judicious decisions are often made without being

“principled” in the pertinent sense. Nonetheless, the resolution of principled

conflicts will frequently appeal to (1) one or more external principles, (2) a

procedure, (3) a form of authority, (4) a balancing of principles, or (5) a specifica-

tion of the principles in conflict. If a procedure or an authority is the best resource

for conflict resolution, this in itself will be determined by principles (or rules) that

designate the procedure or authority. Principles will therefore be integrally

involved in each of these five forms of appeal.

In situations of moral conflict we often need latitude to assess the various moral

demands, leaving room for negotiation and compromise, but such negotiation

and compromise can follow the path of (1) through (5), and so need not be

unprincipled. For example, in some circumstances two or more morally accep-

table alternatives are unavoidably in conflict, because both alternatives present

good reasons for action. Both present good but not decisive or solely sufficient

reasons. Here the best course is often to further specify the precise commitments

of principles, as will be explained later.

Sources of Principles

The four categories of principles are all drawn from the common morality. By

“the commonmorality” I mean the morality that all reasonable persons share and

acknowledge—common-sense ethics, as it is sometimes called. A substantial

social consensus exists about general principles and rules in the common mor-

ality, far more consensus than exists about general norms in philosophical ethical

theories. From this perspective, a paradigm of philosophical theory or method

P R I N C I P L E S A N D O T H E R E M E R G I N G P A R A D I G M S I N B I O E T H I C S 155

should be resisted if it cannot be made coherent with preexistent common-

morality understandings of what John Rawls calls our considered judgments—

that is, those moral convictions in which we have the highest confidence and

believe to have the lowest level of bias, such as principles that prohibit racial

discrimination, religious intolerance, and political favoritism.

Traditional health care contexts often supply more specific moral content than

principles, typically through traditions of role responsibilities. These traditions

supply an understanding of obligations and virtues as they have been adapted over

the centuries for professional practice. The health professional’s obligations,

rights, and virtues have been framed primarily as professional commitments to

shield patients from harm and provide medical care—fundamental obligations of

nonmaleficence and beneficence. Professional dedication to these norms has been

part of the self-understanding of physicians. For example, physicians have tradi-

tionally taken the view that disclosing certain forms of information drawn from

patient records can cause harm to patients under their care and that medical ethics

obligates them to maintain confidentiality.

The principle of nonmaleficence provides perhaps the best example. This

principle has long been associated in medicine with the injunction primum non

nocere: “Above all [or first], do no harm,” a maxim often mistakenly attributed to

the Hippocratic tradition.2 It has an equally prestigious position in the history of

moral philosophy. John Stuart Mill, for example, praised the moral rules of

nonmaleficence as “that which alone preserves peace among human beings.”3

British physician Thomas Percival furnished the first developed account of health

care ethics, in which he maintained that principles of nonmaleficence and benefi-

cence fix the physician’s primary obligations and triumph even over the patient’s

autonomy rights in a circumstance of potential harm to patients:

To a patient . . .who makes inquiries which, if faithfully answered, might


truth. His right to it is suspended, and even annihilated; because, its

beneficial nature being reversed, it would be deeply injurious to himself, to

his family, and to the public. And he has the strongest claim, from the trust

reposed in his physician, as well as from the common principles of humanity,

to be guarded against whatever would be detrimental to him.4

Like the Hippocratics, Percival accepted as the first principle of medical ethics

that the patient’s best medical interest determines the physician’s obligations, and

he conceived the central virtues of the physician through models of benevolence

and sympathetic tenderness, as they serve to promote the patient’s welfare.

Recently the idea has flourished in biomedical ethics that the physician’s moral

responsibility should be understood less in terms of traditional ideals of medical


benefit and more in terms of the patients’ rights of self-determination, including

rights to truthful disclosure, confidentiality, privacy, disclosure, and consent, as

well as welfare rights rooted in claims of justice. These proposals have moved

medical ethics from its traditional preoccupation with a patient-welfare model

toward an autonomy model of the care of patient, while also confronting the field

with a wider set of social concerns, such as the right to health care. For this reason,

principles of autonomy and justice have increased in importance in bioethics.

The justification for choosing the particular four groups of principles that

Childress and I defend is therefore partially rooted historically in medical tradi-

tions of health care ethics and partially rooted in contemporary contexts in which

principles of autonomy and justice point to an important aspect of morality that

was traditionally neglected in health care ethics. However, the common morality

is itself the ultimate source of these basic principles.

The Need for Additional Specification of Principles

To say that principles find support in traditions of health care is not to say that

their appearance in an ethical theory or in a developed paradigm of biomedical

ethics is identical to their appearance in the traditions from which they spring.

Even the common morality falls far short of a well-articulated and specified

paradigm. Conceptual clarification and attempts to increase coherence give

shape and substance to principles, much as judges in their opinions express and

develop the commitments of legal precedents and principles of law to suit the

cases before them.

Both ethical theories and the common morality contain regions of indetermi-

nacy that need reduction through further development of norms in the system,

augmenting them with a more specific moral content. In light of the indetermi-

nacy found in principles and all general norms, I follow Henry Richardson in

arguing that the specification of norms involves filling in details in order to

overcome moral conflicts and the incompleteness of principles and rules.5

Specification is the progressive and substantive delineation of principles and

rules that gives them a more specific and practical content. Because principles

are at a lofty level of abstraction, little practical content can be drawn directly from

them. More precision through specification is therefore essential for regulative

and decision-making contexts.

Principles are not applied but rather are explicated andmade suitable for specific

tasks, often by developing policies. Judgment and decision making are essential

for this interpretive process. For this reason philosophers such as John Mackie

rightly argue that ethics is “invented.” Mackie does not mean that individuals

create either the common morality or personal moral policies, but that what he


calls “intersubjective standards” are creatively built up over time through com-

munal agreements and decision making. What is morally demanded, enforced,

and condemned is less a matter of what is already present in basic principles and

more a matter of what we decide by reference to and in development of those

principles.

As a simple example of specification and invention in this sense, consider

conflicts of obligation that emerge from the dual roles of research scientist and

clinical practitioner. As an investigator, the physician has an obligation to gen-

erate scientific knowledge that will benefit future patients. As a clinical practi-

tioner, the physician has obligations of care that require acting in the best interests

of present patients. The notion of a physician-as-scientist suggests two roles that

pull in different directions, each role having its own specifiable set of obligations.6

How, then, do we make these various obligations more precise, specific, and

coherent when they come into conflict?

One possibility is to segregate the roles so that they cannot conflict—for

example, specifying that physicians with clinical responsibilities cannot use their

own patients when discharging research responsibilities. This formulation is an

“invention” specifying that a physician’s obligations of beneficence to patients

must not be confounded or compromised by research obligations. This specifica-

tion will solve some problems about the dual role, but will leave others unresolved

and in need of additional inventiveness and specification. For example, it might

be in everyone’s best interest in some circumstances for a set of physicians to

assume both roles, although we know that conflicts of interest are present for these

physicians. We might then specify that “Physicians can simultaneously accept

clinical and research obligations for the same patients only if a full disclosure is

made to the patients of the dual role and of any conflicts of interest present in the

dual role.” This specification attempts to make the obligations jointly acceptable

by adding obligations of disclosure that did not previously exist.

Such specifications will at times involve a balancing of principles, at times an

appending of additional obligations, and at other times a development of one or

more principles by making them more precise for purposes of policy. In these

ways we become more specific and practical and maintain fidelity to our original

principle(s). This strategy has the advantage of allowing us to unpack our

evaluative commitments and to expand them as well, presumably achieving a

more workable and a more coherent body of contextually relevant norms. Of

course, many already specified norms will need further specification as new or

unanticipated circumstances arise. All moral norms are potentially subject to this

process of inventive specification, and progressive specification will increasingly

reduce types of conflict and insufficiency of content.7

There are tangled problems about the best method to use in order to achieve

specification and about how we can justify a proposed specification, but it seems


clear that specification is needed, not merely a bare appeal to principles and rules.

The model of analysis for reaching specification and justification in health care

ethics that Childress and I have used is that of a dialectical balancing of principles

against other moral considerations, in an attempt to achieve general coherence

and a mutual support among the accepted norms. A now widely accepted method

of this general description that can be used to help in the process of the specifica-

tion of principles is “reflective equilibrium.” This method was formulated by John

Rawls for use in the construction of a general ethical theory. It views the accep-

tance of principles as properly beginning with considered judgments, and then

requires a matching, pruning, and developing of considered judgments and prin-

ciples in an attempt to make them coherent. We start with paradigms of what is

morally proper or morally improper, and we then search for specifications of

principles that are consistent with these paradigms and consistent with each other.8

A specified principle, then, is acceptable in a system of norms if it heightens the

mutual support of other norms in the system that have themselves survived in

reflective equilibrium. This understanding of the principles paradigm assumes

that no canonical content exists for bioethics beyond the common morality.

There is no scripture, no authoritative interpretation of anything analogous to

scripture, and no authoritative interpretation of that large mass of judgments,

rules, standards of virtue, and the like that we often collectively sum up by use of

words such as “bioethics.” A principle-based account likewise disavows models of

a single ultimate principle of ethics and of absolute rules. The principles approach

supports a method of inventive content expansion intomore specific norms, not a

system layered by priorities among rules or among categories of ethics. From this

perspective, principles grounded in the common morality are the background

framework, but also the point at which the real work of policy development and

moral judgment begin.

ALTERNATIVE PARADIGMS

Several alternative paradigms have arisen in recent years, some of whose propo-

nents have been sharply critical of principles. I will consider three such alter-

natives. Although my aim is chiefly to place criticisms of principles in a proper

perspective, I have a secondary aim. Critics have often appropriately pointed to

limits in the principles paradigm, specifically limits of scope, practicability, and

justificatory power. Much can be learned from this commentary about the points

at which even carefully specified principles are inadequate to provide a compre-

hensive account of the moral life. Alternative paradigms usually exhibit their

primary strength at these points. In my assessments to follow, I will maintain that

these alternatives are congenial to, and not rivals of, a principle-based account.


Casuistry as an Alternative Paradigm

The term casuistry, derived from the Latin casus for “case,” is today used primarily

to refer to a method of using cases to analyze and propose solutions for moral

problems. Clinical bioethics is essentially casuistical, but “casuistry” usually refers

to a specific method of analyzing and generalizing from cases. The casuistical

method is to start with paradigm cases whose conclusions are settled, and then to

compare and contrast the central features in these settled cases with the features of

cases to be decided. Maxims drawn from past cases and specific analogies are used

to support recommendations for new cases. To use an analogy to case law and the

doctrine of precedent, when judicial decisions become authoritative, these deci-

sions have the potential to become authoritative for other judges confronting

similar cases in similar circumstances and with similar facts. Precedents also bind

by restraining the judgments that may be made in new cases.

In casuistical ethics, moral authority proceeds from the settled paradigm cases

and maxims, but there are no rigid rules or principles, because particular circum-

stances and their features alter the way cases are handled and decided. Just as case

law (legal rules) develops incrementally from cases, so the moral law (a set of moral

rules) develops incrementally in casuistry.9 However, moral rules only pick out the

salient features of cases and must be used with caution and discernment. In clinical

medical ethics, for example, prior precedent cases allow us to focus by analogical

reasoning on practical decision making in new cases, but very different conclusions

may be reached in the new cases, depending on their novel features. Characteristic

features of contemporary casuistry include this premium on case interpretation

together with a strong preference for analogical reasoning over theory.10

A Rejection of Principles and Theory

Some contemporary casuists are dissatisfied with principle-based theories, espe-

cially in clinical contexts. They view clinical medical ethics as a discipline arising

from clinical practice, rather than from an application of general ethical principles

to cases.11 However, it is not clear why some contemporary casuists react as

negatively to principles as they often do. In the great Latin traditions of casuistry

emanating from Cicero, students were taught how to use both principles and

analogies from prior cases to propose resolutions in a new case. Both Rabbinical

and Roman common-law traditions of casuistry continued this practice.

Although the sources of contemporary hostility to principles and theory are

difficult to pinpoint, I can briefly treat somemainstream objections. A first type of

objection derives from the close connection some casuists see between principles

and theory, particularly when theory is depicted by its proponents as a unified


theory with impartial and universal principles. The underlying aspiration of such

theories seems, according to these casuists, an emulation of the natural sciences,

by locating what is most general and universal in ethics, expressed in precise

principles that enjoy the high measure of confidence found in scientific principles

or laws. Casuists hold instead that ethics is neither a science nor a theory fash-

ioned along the lines of traditional ethical theories such as utilitarianism and

Kantianism. Rather, ethics is based on seasoned practices rooted in experience.12

Consider an analogy to the way a physician thinks when making a diagnosis and

then a recommendation to a patient. Many individual factors, including the

patient’s medical history, the physician’s successes with similar patients, and

paradigms of expected outcomes, will play a role in formulating a judgment

and recommendation to a patient, which may be very different from the recom-

mendation that will be made to the next patient with the same malady.

A second reason for hostility to principles is that moral philosophers have often

regarded cases as merely a set of facts that can be used to help understand moral

principles and problems, but lacking in the means to resolve the moral problems

presented by new cases. Casuists maintain that when reasoning through cases, we

may legitimately find that we need not appeal to principles, rules, rights, or

virtues. For example, when principles, rules, or rights conflict, and appeals to

higher principles, rules, or rights have been exhausted, we still need to make

reasoned moral judgments. Here the casuist holds that moral reasoning invokes

not principles, but narratives, paradigm cases, analogies, models, classification

schemes, and even immediate intuition and discerning insight.13

Third, when principles are interpreted inflexibly, irrespective of the nuances of

cases, some casuists find the principles “tyrannical,” on grounds that they block

compromise and the resolution of moral problems by generating a gridlock of

conflicting principled stands; moral debate then becomes intemperate. This

impasse can often only be avoided, from the casuists’ perspective, by focusing

on points of shared agreement about cases, not on abstract principles.14

Fourth, casuists maintain that principles and rules are typically too indetermi-

nate to yield specific moral judgments (for reasons already discussed). It is there-

fore impossible that there be a unidirectional movement of thought from principles

to cases. Indeed, specified principles will assume an adequately determinate form

only after reflection on particular cases. The determinate content in practical

principles will therefore, at least in part, be fixed by the authority of the cases.

Fifth, casuists argue that even carefully specified principles must still be weighed

and balanced in accordance with the demands and nuances of particular circum-

stances. An interpreting, weighing, and balancing of principles is essential when-

ever the particular features of cases cannot have been fully anticipated by a prior

process of specification. For example, a physician’s judgment about the decisions a

particular patient should be encouraged to make or discouraged from making are


often influenced by how responsible the physician thinks the patient is, and every

case presents a person at a different level of responsibility. Again, a principle is less

an applied instrument than a part of a wider process of deliberation.

It does not follow from these five critical appraisals that casuists need be hostile

to principles, but only that principles must be interpreted to be coherent with the

casuists’ paradigm of moral reasoning. Reasonable casuists find the gradual

movement from paradigm cases to other cases to be an endeavor that eventuates

in principles, which in turn can be helpful in spotting themorally relevant features

in new cases. Cases can be ordered under a principle through a paradigm and then

extended by analogy to new cases. As abstract generalizations, they help express

the received learning derived from the struggle with cases and capture the con-

nections between cases. However, from the casuists’ perspective, principles, so

understood, are merely summaries of our experience in reflecting on cases, not

norms that are independent of cases.

Some Problems with Casuistry

Although much in these casuistical arguments is acceptable, proponents have

sometimes overstated the promise and power of their account while understating

their reliance on theory and principles. Casuists often write as if cases lead to

moral paradigms, analogies, or judgments by their facts alone. But this claim is

dubious, as the great classical casuists readily acknowledged. The properties that

we observe to be of moral importance in cases are picked out by the values (and

perhaps the theories) that we have already accepted as being morally important.

Consider the following fact, which we might discover in a case: “Person M cannot

survive without person S’s bone marrow.” What shall we conclude morally?

Nothing, a casuist might say, until we know the full range of facts in the case.

But no matter how many facts are stacked one on top of the other, we will still

need some sort of value premise—for example, “Everyone ought to help others

survive through bone marrow transplant donations” in order to reach a conclu-

sion such as, “S ought to donate his bone marrow.” The value premise, which is a

principle or rule, bridges the gap between factual premises and the evaluative

conclusion. Prior to adding this premise, it is not possible to reach the conclusion.

The casuist will face this same problem in every case.

Appeals to “paradigm cases” may mask this fact. Paradigm cases become

paradigms because of prior commitments to central values (and perhaps theories)

that are preserved from one case to the next case, and principles typically play a

legitimate role in determining the acceptability of what is transferred from case to

case. For the casuist to move constructively from case to case, a norm of moral

relevance must connect the cases. Rules of relevant features across cases will not


themselves be merely a part of the case, but a way of interpreting and linking cases.

Even to recognize a case as a paradigm case is to accept whatever principles or

rules allow the paradigms to be extended to other cases.

Jonsen treats this problem by distinguishing descriptive elements in a case from

moral maxims that inform judgment about the case: “These maxims provide the

‘morals’ of the story. For most cases of interest, there are several morals, because

several maxims seem to conflict. The work of casuistry is to determine which

maxim should rule the case and to what extent.”15 So understood, casuistry

presupposes principles (maxims or rules) and takes them to be essential elements

in moral reasoning. The principles are present prior to the decision, and then

selected and weighed in the circumstances. This account agrees with the main

claims in the principles paradigm, not a rival paradigm.

Jonsen notes that he does not dismiss principles even though he diminishes

their importance: “When maxims such as ‘Do no harm,’ or ‘Informed consent is

obligatory,’ are invoked, they represent, as it were, cut-down versions of the major

principles relevant to the topic, such as beneficence and autonomy, cut down to fit

the nature of the topic and the kinds of circumstances that pertain to it.”16 Jonsen

goes on to point out that casuistry is “complementary to principles” in a manner

that still needs to be worked out in moral philosophy. In my view, this language of

“cut-down versions” of principles can be interpreted as meaning “built-up,”

because there is a directed specification of content, not a pruning. Jonsen points

in the same direction that Childress and I do when we use the language of

“specification,” “support,” and the like. Like us, Jonsen sees an intimate connec-

tion between principles and what we call progressive specification to rules (his

maxims) in order to meet the demands of particular contexts. His main fear,

which we share, is that principles will be interpreted inflexibly.

Moral reasoning can again be made analogous to legal reasoning in courts: If a

legal principle commits a judge to an unacceptable judgment, the judge needs to

modify or supplement the principle in a way that renders the judge’s beliefs about

the law as coherent as possible. If a well-founded principle demands a change in a

particular judgment, the overriding claims of consistency with precedent may

require that the judgment be adjusted, rather than the principle.17 Sometimes both

judgments and principles need revision. Either way, principles play a central role.

Casuists also have a problem with moral conflict. Cases are typically amenable

to competing judgments, and it is inadequate to be told that cases extend beyond

themselves and evolve into appropriate paradigms. Cases could evolve in disas-

trous ways because they were improperly treated from the outset by a perilous

analogy. Casuists have no methodological resource (by appeal to case alone) to

prevent a biased development of cases and a neglect of relevant features of cases.

This problem caused the decline of casuistry after 1650, when it became increas-

ingly evident that opposite conclusions could be easily “justified” by competing


casuistical forms of argument. So-called “moral laxity” destroyed classical casu-

istry. The same laxity will doom contemporary casuistry unless it is fortified by

stable principles.

Finally, how does justification occur in casuistry? Given the many different

types of appeal that might be made in any given case (analogies, generalizations,

character judgments, etc.), several different “right” answers can be offered on any

given occasion. This problem exists for virtually all moral theories, and so is not a

problem unique to casuistry. However, without a stable framework of norms,

casuists leave too much room for judgment and have too few resources to prevent

prejudiced or poorly formulated judgments and social conventions.

In the end, casuists seem ambivalent about, but not opposed to, principles. On

the one hand, a limited role is acknowledged. Jonsen says, “This casuistic analysis

does not deny the relevance of principle and theory.”18On the other hand, casuists

denounce firm and firmly held principles as tyrannical, and criticize certain kinds

of appeals to principles as “moralistic” and “not a serious ethical analysis.”19

Proponents of casuistry seem most deeply worried not about a reasonable use of

reasonable principles, but excessive forms of reliance in recent philosophy on

universal principles. It is, then, incorrect to make an account based on principles a

rival of casuistry. Casuists rightly point to the gap that exists between principles

and good decision making, but their account will fall victim to the same charge if

it leaves a similar gap between cases and good decision making.

Virtue Theory as an Alternative Paradigm

Recent ethical theory has returned to another prominent classical paradigm:

character and virtue. This paradigm has been used to pose a challenge to prin-

ciple-based theories, which typically attend to actions and obligations rather than

agents and their virtues. The language of principles and obligations, some virtue

theorists claim, descends from assessments of virtue, character, and motives.20

Major writers in the virtue tradition have held that, to cite an observation of

Hume’s, “If a man have a lively sense of honour and virtue, with moderate

passions, his conduct will always be conformable to the rules of morality; or if

he depart from them, his return will be easy and expeditious.”21 Various writers in

biomedical ethics have adopted this perspective, while arguing that the attempt in

an obligation-oriented account to make principles, rules, codes, or procedures

paradigmatic will result in worse rather than better decisions and actions, because

the only reliable protection against unacceptable ethical behavior is virtuous

character.22 From this perspective, character is more important both in institu-

tions and in personal encounters than is conformity to principles.


This line of argument has merit, but needs to be buttressed by a more careful

statement of the nature of the virtues and their connection to principles. A moral

virtue is a trait of character valued for moral reasons. Virtue requires properly

motivated dispositions and desires when performing actions, and therefore is not

reducible to acting in accordance with or for the sake of principles of obligation.

We care morally about a person’s motivation, and we care especially about

characteristic forms of motivation. Persons motivated by compassion and per-

sonal affection meet our approbation when others who act the same way but from

different motives would not. For example, imagine a physician who meets all of

his moral obligations but whose underlying motives and desires are morally

inappropriate. This physician cares not at all about being of service to people or

creating a better environment in his office. He only wants to make money and

avoid malpractice suits. Although this man meets his moral obligations, his

character is deeply defective. The admirable compassion guiding the lives of

many dedicated health professionals is absent in this person, who merely engages

in following the socially required principles and rules of behavior.

Properly motivated persons do not merely follow principles and rules; they

have a morally appropriate desire to act as they do. One can be disposed by habit

to do what is right in accordance with the demands of principles yet be inappro-

priately motivated. To speak of a good or virtuous action done from principle is

usually elliptical for an evaluation of the motive or desire underlying the action.23

For example, if a person’s act of benefiting another person is to elicit moral praise,

the person’s motive must be to benefit; it cannot be a motive such as the desire to

be rewarded for supplying the benefit. Right motive is essential for virtue, and a

virtuous character is constituted by an appropriate motive or motivational

structure. Persons who characteristically perform morally right actions from

principles without a right set of motives and desires are not morally virtuous,

even if they always perform the right action from the right principle.

This paradigm of the moral person succeeds in addressing the moral worth of

persons more adequately than does a principle-based theory of right action. The

paradigm appropriately indicates that virtue cannot be reduced to right action in

accordance with principles or rules. Kindness, for example, cannot be reduced to a

rule-structured action or precept, as if kindness were a matter of following a

recipe. Kindness is a disposition to treat people in certain ways from specific

motives and desires. We are often more concerned about these motives and

desires in persons than about the conformity of their acts to rules. For example,

when a physician takes care of us, we expect his or her actions to be motivated

from a sense of principled obligation to us, but we expect more as well. We expect

the physician to have a desire to take care of us and to want to maintain our hope

and keep us from despair. The physician or nurse who acts exclusively from

principles may lack the virtue of caring that is implied by the term medical care.


Absent this virtue, the physician or nurse is morally deficient. Accordingly, to look

at principled actions without also looking at virtues is to miss a large segment of

the moral life.

These arguments in defense of virtue ethics are entirely compelling, but giving

the virtues a central place in the moral life does not indicate that a virtue-based

paradigm should bump or be granted priority over a principle-based paradigm.

The two approaches have different emphases, but they can easily be mutually

reinforcing if one takes the view that ethical theory is richer and more complete

if the virtues are included. The strength of the virtue paradigm is in the central

role played by the motivational structure of a virtuous person, which often is as

serviceable in guiding actions as are rules and principles.24 But the actions of

persons with a virtuous character are not morally acceptable merely because

they are performed by a person of good character. People of good character can

perform improper actions because they have incorrect information about con-

sequences, make incorrect judgments, or fail to grasp what should be done. We

sometimes cannot evaluate a motive, a moral emotion, or a form of expression

as appropriate or inappropriate unless we have some basis for the judgment that

actions are obligatory, prohibited, or permissible. It is, therefore, doubtful that

virtue ethics can adequately explain and justify many assertions of the rightness

or wrongness of actions without resort to principles and rules.

If we rely, as we should, on traits of character such as sympathy and benevo-

lence to motivate us morally, we should also be prepared for our motives to be

partial and to need correction by impartial moral principles. For example, we are

likely to judge persons more favorably as they are close to us in intimate relation-

ships. Yet on some occasions those who are distant from us deserve to be judged

more favorably than we are disposed to judge them. Virtues, then, do not preclude

impartial principles and rules, and they often work together seamlessly. In addi-

tion, virtues such as wisdom and discernment involve understanding both that

and how principles and rules are relevant in a variety of circumstances. Principles

and virtues are in this respect congenial. Both require attention and sensitivity

attuned to the demands of a particular context. Respect for autonomy and

beneficence will be as varied in different contexts as compassion and discernment,

and the ways in which health professionals manifest these principles and virtues in

the care of patients will be as different as the ways in which devoted parents care

for their children.

Many virtues dispose persons to act in accordance with principles and rules, and

a person’s virtuous character is often found in a practical understanding of how to

employ a principle in a particular case through a creative response in meeting

responsibilities. The ability to understandwhat needs to be done for patients, as well

as how to do it, and then acting with sensitive and caring responses are moral

qualities of character, not merely forms of practical intelligence and judgment.


These forms of caring sometimes open up discerning insights into what is at stake,

what counts the most, and what needs to be done. At the same time, even a virtue

such as moral integrity, which accommodates a wide variety of moral beliefs, is

often a matter primarily of living in fidelity to one’s moral norms and judgments.

Moral integrity in science, medicine, and health care should be understood pre-

dominately in terms of fidelity to principles and rules that can be identified in the

common morality and in the traditions of health care. Many other virtues, such as

conscientiousness, could be similarly treated in terms of a serious commitment to

follow principles and rules.

Finally, it deserves notice that some areas of the moral life are not readily

frameable or interpretable in the language of virtue theory. Committee reviews in

hospitals and research centers provide a typical case in contemporary bioethics.

When strangers meet in professional settings, character judgments will often play

a less significant role than norms that express rights and appropriate procedures.

The same is true in the enforcement of institutional rules and in framing public

policy. Dispensing with specified principles and rules of obligation in these

settings would be an unwarranted loss in the moral life.

In his work on virtue ethics, Edmund Pellegrino has observed that “Today’s

challenge is not how to demonstrate the superiority of one normative theory over

the other, but rather how to relate each to the other in a matrix that does justice to

each and assigns each its proper normative force.”25 I quite agree. I have tried only

to show that principles can provide a defensible normative framework that fits

comfortably in bioethics with the virtue theory tradition.

Dartmouth Descriptivism and the Critique of Principlism

Not everyone who accepts norms of obligation and moral ideals agrees that

principles provide the best framework for health care ethics. At the head of this

line of critics are the self-described Dartmouth descriptivists—K. Danner

Clouser, Bernard Gert, and Ronald Green. At the base of this account is a

theory of “morality,” a term that references the entire “moral system” of rules,

moral ideals, and procedures for determining when it is justified to override or

violate a moral rule. Paradoxically, Clouser and Gert are probably as close to

Childress and me in their conception of the content of morality and philosophical

method as they are to any other writers in biomedical ethics. The problem is that,

although they accept general rules of obligation, they do not believe that general

principles provide a reliable framework. They look to more specific rules, ideals,

and procedures arranged in a structured system as the proper surrogate.

They refer to the account that Childress and I have developed as “princip-

lism.”26 Clouser and Gert bring the following accusations against systems of


general principles: (1) principles are little more than checklists or headings

for values and have no deep moral substance that can guide practice in the

way moral rules do; (2) analyses of principles fail to provide a theory of

justification or a theory that ties the principles together, with the consequence

that principles are ad hoc constructions lacking systematic order; and (3) prima

facie principles often compete in difficult circumstances, yet the underlying

philosophical theory is too weak both to decide how to adjudicate the conflict

in particular cases and to deal theoretically with the problem of a conflict of

principles.

These problems are worthy of careful and sustained reflection in moral theory.

However, I doubt that Dartmouth descriptivists themselves have surmounted the

problems they lay at the door of principle-based approaches. The primary differ-

ence between what Childress and I call principles and what they call rules is that

their rules tend in their abstract formulation to have a more directive and specific

content than our principles, thereby at first glance giving more guidance in the

moral life. But we have pointed out this same fact since our first edition (in 1979).

We have always insisted that turning general principles into specific rules is

essential for health care ethics. Mere unspecified principles will not take us very

far in bioethics.

There is also not more and not less normative content in their rules and

ours, and neither more nor less direction in the moral life. It is true that

principles function to order and classify as much as to give prescriptive

guidance, and therefore principles do serve a labeling and organizing function.

However, this feature only indicates again that principles are abstract starting

points in need of additional specification. Clouser and Gert misleadingly

suggest that principles only sort and classify without giving significant norma-

tive guidance. Logically, the function of principles is to guide conduct, and in

pure cases free of conflicting obligations, principles often provide the needed

guidance without additional rules. But principles are not stateable with an eye

to eliminating the many possible conflicts among principles, because no

system of guidelines (principles or rules) could reasonably anticipate the full

range of conflicts.

Moreover, a set of rules almost identical to the Clouser-Gert rules is already

included in the account of principles and rules that Childress and I propose.

We have maintained that principles lend support to more specific moral

rules, including those accepted by Clouser and Gert, and that more than one

principle—respect for autonomy and nonmaleficence, say—may support a

single rule (e.g., a rule of medical confidentiality). The following is a compar-

ison between a sample of the rules we defend under the heading of the principle

of nonmaleficence and a directly related sample of basic moral rules defended

by Gert and Clouser:


Beauchamp & Childress Gert & Clouser

4 Rules Based on Nonmaleficence 4 of the 10 Basic Rules

1. Do not kill. 1. Don’t kill.

2. Do not cause pain. 2. Don’t cause pain.

3. Do not incapacitate. 3. Don’t disable.

4. Do not deprive of goods. 4. Don’t deprive of pleasure.

No substantive moral difference distinguishes these sets of rules. The method of

deriving and supporting the rules is different in our accounts, but at present I am

claiming only that their rules either do not or need not differ in content from ours

and that their rules are no more specific and directive than ours.

Second, Gert and Clouser are critical of us for making both nonmaleficence and

beneficence principles of obligation. They maintain that there are no moral rules

of beneficence that state obligations. Our only obligations in the moral life, apart

from duties encountered in roles and other stations of duty, they regard as

captured by moral rules of nonmaleficence, which prohibit causing harm or

evil. Their reason is that the goal of morality is the minimization of evil, not the

promotion of good. Rational persons can act impartially at all times in regard to

all persons with the aim of not causing evil, but they cannot impartially promote

the good for all persons at all times. As Clouser states the view, the moral rules all

admonish us to avoid harming and are all prohibitions, but, by contrast, moral

ideals are only to be encouraged. Since beneficence can only be encouraged and

not required, it is a moral ideal only.27

This thesis is neithermorally correct nor supported within the account of moral

obligations presented by Gert and Clouser. The implication of this claim is that

one is never morally required (i.e., obligated by moral rules) to prevent or remove

harm or evil, but only to avoid causing it. There is no requirement to do anything,

only to avoid causing harmful events. Childress and I believe this thesis misreads

common morality. The following cases of what one is not obligated to do in the

official Clouser-Gert account of obligations of beneficence point to why their view

is doubtful:

1. Mr. X’s life is in great peril, but if I warn him through a phone call of the

peril, he will be fine. I am not obligated to make the phone call.

2. A young toddler has wandered onto a busy street, having become sepa-

rated from his mother. I can save his life simply by picking him up. It

would be nice of me to do so, but I am not obligated to do so. After all,

there are no obligations, ever, to prevent harm, apart from roles that fix

obligations for us. A policeman would be obligated to lift the child from

the street, but not I.


3. A blind man on the street has obviously lost his way, and he asks me for

directions. I am not obligated to provide the directions.

4. I am the only witness to a serious automobile accident that has left the

drivers unconscious. I am not obligated to stop or to pick up my car

telephone and call an ambulance as I speed on my way.

These examples suggest that there are moral obligations of beneficence and that

they should be included in any system whose goal is to capture the nature and

scope of morality. If Clouser and Gert believe that we are never under such

obligations and that a morality of obligations does not contain at least as much

beneficence as these four examples suggest, then we deeply disagree about the

content of morality. However, as Gert points out in his book, Morality: A New

Justification of the Moral Rules, which Clouser acknowledges to be “the basis” of

his understanding of method and content in ethics, Gert does believe that one is

morally obligated to act in circumstances precisely like those in my four examples.

His reason is his acceptance of the general rule “Do your duty,” which he gives an

interpretation broad enough to incorporate the principle of beneficence. He

explains his system and its commitments as follows:

Although duties, in general, go with offices, jobs, roles, etc., there are some

duties that seem more general. . . . A person has a duty . . . because of some

special circumstances, for example, his job or his relationships. . . . In any

civilized society, if a child collapses in your arms, you have a duty to seek

help. You cannot simply lay him out on the ground and walk away. In most

civilized societies one has a duty to help when (1) one is in physical proximity

to someone in need of help to avoid a serious evil, usually death or serious

injury, (2) one is in a unique or close to unique position to provide that help

and (3) it would be relatively cost-free for one to provide that help.28

Although Gert insists that these requirements are all supported by the moral rule

“Do your duty,” they are effectively identical to those obligations that follow from

what Childress and I call—following moral tradition in the eighteenth, nineteenth,

and twentieth centuries—beneficence. It therefore cannot be the case that in Gert’s

system there are no obligations of beneficence in our sense of beneficence. Gert and

Clouser cannot be criticizing our views about rules of beneficence (though they

could object to how we ground and justify such rules), because we are in effect in

agreement on all the substantive issues about what is morally required. Often when

Clouser and Gert critique our views it appears that they want to categorize all

obligations of beneficence as moral ideals, but it would be inconsistent to take this

line about our account of beneficence, given the latent (but deep) commitments to

beneficence in Gert’s moral theory.


To generalize, much in principlism that Clouser and Gert appear to reject can

be situated in their account under Gert’s final rule, “Do your job” (or “Don’t

avoid doing your job”). If this interpretation is correct, our theories are far more

compatible than they allow, though it is always worth bearing in mind that

Childress and I are defending only a professional ethics, not a general moral

theory, as Gert is. Still, it is difficult to see how the impartial rule theory of Clouser

and Gert provides a real alternative to our substantive claims about the nature and

scope of obligations. I have not here discussed methodology, but our methods are

also, in many respects, compatible. Gert and Clouser therefore seem more like

congenial partners than hostile rivals.

The major difference between our theory and theirs has nothing to do with

whether principles or rules are primary or secondary normative guides in a theory,

but rather with several aspects of their theory that I would reject. First, they assume

that there is, or at least can be, what they call a “well-developed unified theory” that

removes conflicting principles and consistently expresses the grounds of correct

judgment—in effect, a canon of rules and theory that expresses the “unity and

universality of morality.” They fault us heavily for believing that more than one kind

of ethical theory can justify a moral belief and insist that we must do the theoretical

work of showing the basis of principles. They insist that to avoid relativism there can

only be “a single unified ethical theory” and that there cannot be “several sources of

final justification.”29 I reject each of these claims, but at the same time I recognize

them as reasonable philosophical requests for further argument.

In the end, I think my general conception of morality and its primary elements

is notably similar to Gert’s moral system.We both accept the view that morality is

an informal public system of norms with authority to guide the actions of all

persons capable of moral agency in their behavior that affects others, and we agree

that the goal of morality is to prevent or limit problems of harm, conflict,

hostility, and the like through the most suitable moral norms for doing so.

CONCLUSION

I have argued that the moral universe should not be divided into rival and

incompatible theories that are principle-based, virtue-based, rights-based, case-

based, rule-based, and the like. Impartial rule theory, casuistry, and virtue

ethics can all be interpreted as consistent with rather than rivals of a prin-

ciple-based account.

In the moral life we often coherently fuse appeals to principles, rules, virtues,

analogies, precedents, and parables. To assign priority to one paradigm of bio-

medical ethics is a suspicious project, and I have not argued that the principles

paradigm is somehow more serious or more worthy than other paradigms. Even


theories with a single ultimate principle, such as utilitarianism and Kantianism,

deserve careful attention for what they can teach us about moral reasoning and

moral theory. The more general (principles, rules, theories, etc.) and the more

particular (feelings, perceptions, case judgments, practices, parables, etc.) should

be coherently united in the moral life and in moral philosophy, not ripped from

their natural habitat and segregated into distinct and rival species.

A careful analysis and specification of principles is consistent with a wide variety

of types of ethical theory, including virtue theory and some accounts that came to

prominence only recently, such as communitarian theories, casuistical theories, and

the ethics of care. Many authors in biomedical ethics continue tomake amistake by

addressing the field as if a principle-based approach is a one-sided and exclusionary

approach to bioethics. At the same time, the principles paradigm must address the

fact that principles are initially attractive because they offer an impartial instrument

to resolve our moral dilemmas, but in concrete circumstances conflicts among the

principles often generate dilemmas rather than resolving them. A defender of

principles will therefore be grateful for help from any resource that can blunt or

reduce intractable dilemmas or disagreements. Every reasonable, insightful, and

useful strategy is one we can ill afford to reject if we are to successfully handle the

diverse set of issues needing treatment in contemporary bioethics.

Notes

1. For example, the National Commission for the Protection of Human Subjects struggled

with principles from 1974 to 1978 because of a mandate from the U.S. Congress requiring that

the National Commission investigate the ethics of research and explore the “basic ethical

principles” of research ethics. The National Commission ultimately developed an abstract

schema of basic principles related to the subject areas of research ethics to which they

primarily apply: Respect for persons applies to informed consent; beneficence applies to

risk–benefit assessment; and justice applies to selection of subjects. In light of this schema, a

general strategy was devised for handling problems of research ethics. See National

Commission for the Protection of Human Subjects of Biomedical and Behavioral

Research, The Belmont Report (Washington, DC: DHEW Publication No. OS 78-0012, 1978).

2. Albert R. Jonsen, “Do No Harm: Axiom of Medical Ethics,” in Philosophical and

Medical Ethics: Its Nature and Significance, ed. Stuart F. Spicker and H. Tristram

Engelhardt, Jr. (Dordrecht: D. Reidel, 1977), 27–41.

3. John Stuart Mill, Utilitarianism, in vol. 10 of the Collected Works of John Stuart Mill

(Toronto: University of Toronto Press, 1969), ch. 5.

4. Thomas Percival, Medical Ethics; or a Code of Institutes and Precepts, Adapted to the

Professional Conduct of Physicians and Surgeons (Manchester, England: S. Russell, 1803),

165–166. Percival’s work served as the pattern for the American Medical Association’s

(AMA) first code of ethics in 1847.


5. Henry S. Richardson, “Specifying Norms as a Way to Resolve Concrete Ethical

Problems,” Philosophy and Public Affairs 19 (Fall 1990): 279–310.

6. See Benjamin Freedman, “Equipoise and the Ethics of Clinical Research,” New

England Journal of Medicine 317 (July 16, 1987): 141–145.

7. See Richardson, “Specifying Norms,” 294.

8. Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), 20ff, 46–

49, 195–201, 577ff.

9. John D. Arras, “Getting Down to Cases: The Revival of Casuistry in Bioethics,”

Journal of Medicine and Philosophy 16 (1991): 31–33; Albert Jonsen and Stephen Toulmin,

The Abuse of Casuistry (Berkeley: University of California Press, 1988), 16–19, 66–67; Jonsen,

“Casuistry and Clinical Ethics,” Theoretical Medicine 7 (1986): 67, 71; and “Casuistry as

Methodology in Clinical Ethics,” Theoretical Medicine 12 (December 1991): 298.

10. Cf. John D. Arras, “Getting Down to Cases,” and “Common LawMorality,”Hastings

Center Report 20 (July/August 1990): 35–37. Arras makes the following constructive recom-

mendations after criticizing casuistry: Use real and lengthy cases, present complex

sequences of cases in which they build on one another, and be cautious regarding the

limits of casuistical methods.

11. Cf. David C. Thomasma, “Why Philosophers Should Offer Ethics Consultations,”

Theoretical Medicine 12 (1991): 129–140.

12. Although some casuists are critical of theory, others encourage principles and theory

construction. See Baruch Brody, Life and Death Decision Making (New York: Oxford

University Press, 1988), 13. For a very different view, see Albert R. Jonsen, “Practice Versus

Theory,” Hastings Center Report 20 (July/August 1990): 32–34. Brody defends theory con-

struction; Jonsen challenges the presumption that “theory is an inseparable companion to

practice” and contrasts theory construction and the art of practice found in casuistry.

13. Jonsen and Toulmin, Abuse of Casuistry, 11–19, 66–67, 251–254, 296–299; Jonsen,

“Casuistry as Methodology in Clinical Ethics,” 299–302; Brody, Life and Death Decision

Making, 12–13, 15n; Arras, “Getting Down to Cases,” 31–33; Jonsen, “Casuistry and Clinical

Ethics,” Theoretical Medicine 7 (1986), 67, 71.

14. Toulmin, “The Tyranny of Principles,”Hastings Center Report 11 (December 1981): 31–39.

15. Jonsen, “Casuistry as Methodology in Clinical Ethics,” 298.

16. Albert R. Jonsen, “Casuistry: An Alternative or Complement to Principles?” Kennedy

Institute of Ethics Journal 5 (1995): 246–247.

17. Cf. Joel Feinberg, Social Philosophy (Englewood Cliffs, NJ: Prentice-Hall, 1973), 34.

18. Jonsen, “Case Analysis in Clinical Ethics,” The Journal of Clinical Ethics 1 (1990): 65.

See Abuse of Casuistry, 10.

19. Jonsen, “American Moralism and the Origin of Bioethics in the United States,” esp.

117, 125–128.

20. See Philippa Foot, Virtues and Vices (Oxford: Basil Blackwell, 1978), and Gregory

Trianosky, “Supererogation, Wrong-doing, and Vice,” Journal of Philosophy 83 (1986):

26–40.

21. David Hume, “The Sceptic,” in his Essays, Moral, Political, and Literary, ed. Eugene

Miller (Indianapolis: LibertyClassics/Liberty Fund, 1985), 169 (par. 29 in the essay).


22. A classic treatment is H. K. Beecher, “Ethics and Clinical Research,” New England

Journal of Medicine 274 (1966): 1354–1360. See also Gregory Pence, Ethical Options in

Medicine (Oradell, NJ: Medical Economics Co., 1980), 177.

23. This formulation is indebted to David Hume, A Treatise of Human Nature, 2nd

edition, ed. L. A. Selby-Bigge and P. H. Nidditch (Oxford: Clarendon Press, 1978), 478; and

John Mackie, Hume’s Moral Theory (London: Routledge and Kegan Paul, 1980), 79–80.

24. See David Solomon, “Internal Objections to Virtue Ethics,” Midwest Studies in

Philosophy 13 (1988): 439.

25. Edmund Pellegrino, “Toward a Virtue-Based Normative Ethics for the Health

Professions,” Kennedy Institute of Ethics Journal 5 (1995): 253–277, esp. 273. See, for more

extensive treatment, Pellegrino’s book (coauthored with David Thomasma), The Virtues in

Medical Practice (New York: Oxford University Press, 1993).

26. See Clouser and Gert, “A Critique of Principlism,” The Journal of Medicine and

Philosophy 15 (April 1990): 219–236; and Ronald M. Green, Gert, and Clouser, “The Method

of Public Morality versus the Method of Principlism,” The Journal of Medicine and

Philosophy 18 (1993). A diverse, but less focused, set of criticisms is found in Ron P.

Hamel, Edwin R. DuBose, and Laurence J. O’Connell, Beyond Principlism (Chicago:

Trinity Press International, 1993).

27. K. Danner Clouser, “Common Morality as an Alternative to Principlism,” Kennedy

Institute of Ethics Journal 5 (1995): 219–236, esp. 225–226, 231.

28. Bernard Gert, Morality: A New Justification of the Moral Rules (New York: Oxford

University Press, 1988), 154–155.

29. Clouser andGert, “A Critique of Principlism,” 231–232; Green, “Method in Bioethics:

A Troubled Assessment,” The Journal of Medicine and Philosophy 15 (1990): 179–197.


10

A DEFENSE OF THE COMMON MORALITY

Phenomena of moral conflict and disagreement have led writers in ethics to two

antithetical conclusions. Some writers maintain that there are objective, universal

moral standards. Others reject all objectivity and universality. To maintain, as do

most writers in bioethics, that there are valid moral distinctions between just and

unjust actions, merciful and unmerciful actions, and humane and inhumane

actions is either to claim that these conclusions hold universally or that they hold

relative to a particular and contingent moral framework. If judgments are entirely

relative to a local framework, then they cannot be applied with universal validity.

I maintain that we can consistently deny universality to some justified moral

norms while claiming universality for others. I argue for this conclusion by locating

universality in the common morality and nonuniversality in other parts of the

moral life, which I call particular moralities. I defend the thesis that there are

universal moral standards through (1) a theory of the objectives of morality, (2)

an account of the norms that achieve those objectives, and (3) an account of

normative justification (both pragmatic and coherentist). I discuss how the

common morality is progressively made specific and the sense in which moral

change occurs.

Several instructive articles have recently been critical of my previously pub-

lished views on the common morality. I will respond here to articles by David

DeGrazia and Leigh Turner, both of whom take “common morality” to refer to a

broader and quite different body of norms than I do.

THE NATURE AND OBJECTIVES OF THE COMMON MORALITY

I understand the common morality as the set of norms shared by all persons

committed to the objectives of morality. The objectives of morality, I will argue,

are those of promoting human flourishing by counteracting conditions that cause

the quality of people’s lives to worsen.

The Nature of the Common Morality

The commonmorality is not merely amorality that differs from othermoralities.1

It is applicable to all persons in all places, and all human conduct is to be judged by

its standards. Virtually all people in all cultures grow up with an understanding of

the basic demands that morality makes upon everyone. They know not to lie,

steal, break promises, and the like. The following are examples of standards of

action (rules of obligation) in the commonmorality: (1) Don’t kill, (2) Don’t cause

pain or suffering to others, (3) Prevent evil or harm from occurring, (4) Rescue

persons in danger, (5) Tell the truth, (6) Nurture the young and dependent,

(7) Keep your promises, (8) Don’t steal, (9) Don’t punish the innocent, and

(10) Treat all persons with equal moral consideration.

The common morality contains standards other than principles of obligation.

Here are 10 examples ofmoral character traits (virtues) recognized in the common

morality: (1) nonmalevolence, (2) honesty, (3) integrity, (4) conscientiousness,

(5) trustworthiness, (6) fidelity, (7) gratitude, (8) truthfulness, (9) lovingness, and

(10) kindness. My claim with respect to these virtues is that they are universally

admired traits of character, that a person is universally recognized as deficient in

moral character if he or she lacks such traits, and that those negative traits that are

the opposite of these virtues and that we call vices (malevolence, dishonesty, lack

of integrity, cruelty, etc.) are substantial moral defects, universally so recognized.

I will hereafter say no more about the virtues and vices. This area of the moral life

is not my concern in this paper; instead, I focus on norms of action.

The Objectives of Morality

Centuries of experience have demonstrated that the human condition tends to

deteriorate into inconvenience, misery, violence, and distrust unless norms of the

sort I listed earlier—the norms of the common morality—are observed. When

complied with, these norms lessen human misery and preventable death. The

object of morality is to prevent or limit problems of indifference, conflict,

hostility, scarce resources, limited information, and the like.


It is an overstatement to maintain that all of the norms that I listed previously

are necessary for the survival of a society (as diverse philosophers and social

scientists have maintained2), but it is not too much to claim that these norms

are necessary to ameliorate or counteract the tendency for the quality of people’s lives

to worsen or for social relationships to disintegrate.3 In every well-functioning

society norms are in place to prohibit lying, breaking promises, causing bodily

harm, stealing, fraud, the taking of life, the neglect of children, and failures to keep

contracts. These norms occupy a central place in the moral life because they have

proven that they successfully achieve the objectives of morality. This success in the

service of human flourishing accounts for their moral authority.

PARTICULAR MORALITIES AS NONUNIVERSAL

I shift now to analysis of particular moralities. Turner states that I do not allow for

pluralism or for the local character of “moral worlds.” However, he misconstrues

my view. Many moral norms are particular to cultures, groups, and even indivi-

duals. Whereas the common morality contains only general moral standards that

are conspicuously abstract, universal, and content-thin, particular moralities

often present concrete, nonuniversal, and content-rich norms. These moralities

implement the responsibilities, aspirations, ideals, attitudes, and sensitivities that

spring from cultural traditions, religious traditions, professional practice, institu-

tional rules, and the like. In some cases, explication of the values in these

moralities requires a special knowledge and may involve refinement by experts

or scholars—as, for example, in the body of Jewish moral norms in the Talmudic

tradition. There may also be full-bodied moral systems set up to adjudicate

conflicts and provide methods for the treatment of borderline cases—for

example, the norms and methods in Roman Catholic casuistry.

Professional moralities, includingmoral codes and traditions of practice, are one

type of particular morality. These moralities may legitimately vary in the way in

which they handle conflict of interest, protocol review, advance directives, and

many other subjects. Moral ideals provide another instructive example of parti-

cular moralities. Following ideals is not morally obligatory, but ideals are uni-

versally admired aspirations.4 Actions done from these ideals are morally good

and praiseworthy, and those who fulfill their ideals can be praised and admired,

but they cannot be blamed or disdained by others if they fail to fulfill aspirational

ideals. Ideals are particularized in local moralities.

Turner states that I defend an ahistorical and an a priori account of morality. It

should be clear from my comments on particular moralities that I do not defend

this view. Particular moralities develop historically. I am also a historicist and

conventionalist in regard to the common morality. However, a defense of this

A D E F E N S E O F T H E C OMM O N MO R A L I T Y 177

position, which is Hobbesian and Humean in inspiration, would require a more

extended discussion than can be undertaken here. Two clarifications must suffice:

First, Turner opines that pluralists form a “third theoretical camp” that is sharply

distinguished from relativists and universalists. I reject this conclusion. A pluralist

who repudiates universal norms is simply one form of relativist. Turner appears to

be a pluralist and a relativist, though I am unsure whether his type of relativism is

normative or nonnormative. He has no means to rebut the thesis that all com-

munally initiated systems of norms are on an equally satisfactory moral footing,

regardless of the principles that underlie them, the reasons for adopting them, or

the consequences of their adoption. Second, Turner maintains that there are, at

local levels (by contrast to the universal level), “shared, historically emergent

understandings and common accounts of moral practices.” This apparently

empirical thesis has both wide and narrow applications. If there can be such

commonly accepted practices locally (in cultures, tribes, religious traditions,

professional organizations, and the like), then there can be commonly accepted

practices universally. Turner and I ought not to disagree about this. Any difference

between us should have to do with how to test empirically for common accep-

tance. I turn, then, to the problem of empirical assumptions and evidence.

THE PROBLEM OF EMPIRICAL JUSTIFICATION

Turner is tenaciously critical of what he sees as the empirical assumptions that

I and my coauthor James Childress make in discussing the common morality. He

says that scant anthropological or historical evidence supports the thesis that a

universal common morality exists. Part of the problem is that Turner includes

under “morality” many more norms—for example, norms that structure social

hierarchies, stations, and gender roles—that I do not include. As best I can tell,

Turner altogether ignores and perhaps rejects the distinction that I make between

common morality and particular moralities.5

In principle, scientific research could prove me—or Turner—wrong. However,

before such research is undertaken, we need to be clear about the concept, or

hypothesis, being tested. For purposes of empirical investigation, my claim is that

all persons committed to morality, and all well-functioning societies, adhere to

the general standards of action enumerated previously. My claim is not, as Turner

seems to believe, that all of the moral norms of all societies are indistinguishable.

What we now know or could know empirically, and whether the propositions that

I have advanced are falsifiable by scientific investigation, are nuanced questions

that are not as straightforward as Turner proposes. It would be difficult to design

empirical studies without either missing the target (i.e., all and only those who are

committed to the objectives of morality) or begging the question. The question


could be begged either by (1) designing the study so that the only persons tested

are those who already have the commitments and beliefs the investigator is testing

for or (2) designing the study so that all persons are tested whether or not they are

committed to the objectives of morality. The first design risks biasing the study in

favor of my hypothesis; the second design risks biasing the study against my

hypothesis.

I have defined the common morality in terms of “the set of norms shared by all

persons committed to the objectives of morality.” Since persons not committed6

to the objectives of morality are not within the scope of my argument, they could

not appropriately be included as subjects in an empirical study. Their beliefs are

not useful for testing my empirical hypothesis. Some, including Turner, will

conclude that I have constructed a circular and self-justifying position; that is,

they will say that I am defining the common morality in terms of a certain type of

commitment and then allowing only those who accept the kinds of norms that I

have identified to qualify as persons committed to the objectives of morality. In

publications referenced by Turner, I have claimed that amoral people, immoral

people, and people driven by ideologies that override moral obligations are not

pursuing the objectives of morality, whatever else they may be pursuing.

I appreciate that this position risks stipulating the meaning and content of

“morality.” Nonetheless, I think the position I defend is the correct one. A full

defense of this position would require a justification of all of the elements of my

account—in particular, the object of morality, considered judgments, the role of

coherence, pragmatic justification, and specification. Here I can do no more than

note three reasons why Turner’s appeals to empirical evidence are unconvincing.

First, no empirical studies known to me show that only some cultural moralities

accept, whereas others reject, the several examples of standards of action that

I earlier submitted as rules of obligation in the common morality. Empirical

investigations of morality concentrate on differences in the way such rules are

embedded in different cultures. These studies assume rather than question the

most general ethical standards. Their results show differences in the interpretation

and specification of shared standards; they do not show that cultures reject basic

common-morality norms. For example, empirical studies do not test whether a

cultural morality includes or rejects rules against theft, promise breaking, or killing.

Instead, investigators study when theft, promise breaking, and killing are deemed in

these cultures to occur, how cultures handle exceptive cases, different conceptions

of killing, and the like. Empirical data show variation in what I refer to as particular

moralities and in specification of the rules of the common morality. These data do

not provide evidence that a common morality does not exist.

Second, the conclusions that I reach can be tested empirically, but as yet

appropriate hypotheses have not been tested. Investigation would center on

persons who had already been screened to ensure that they are committed to


the objectives of morality, but not screened to determine which particular norms

they believe to be the best means to the achievement of those objectives. That is,

persons not committed to the objectives of morality would be excluded from the

study, and the purpose of the study would be to determine whether cultural or

individual differences emerge over the (most general) norms believed to best

achieve the objectives of morality. Should it turn out that those studied do not

share the norms that I hypothesize to comprise the common morality, then there

is no commonmorality of the sort I claim, andmy hypothesis has been falsified. If

norms other than the ones I have specified were demonstrated to be shared across

cultures, this finding would constitute evidence of a commonmorality, albeit one

different from the account I have proposed. Only if nomoral norms were found in

common across cultures would the general hypothesis that a common morality

exists be rejected. Of course, whatever is established about the existence of a

common body of norms, nothing follows about whether those norms are justifi-

able, adequate, in need of change, and the like. I will later consider how this

normative question is distinct from all empirical problems.

Third, and paradoxically, Turner’s notion of “shared, historically emergent

understandings” and “widely shared understandings of moral practices and

policies” errs in the direction of presuming more shared agreement than actually

exists. I agree with the leading thesis in Turner’s paper—“We can find both zones

of consensus and zones of conflict”—but even within local regions of consensus,

there is sure to be dissent and controversy. Consider Turner’s statement that

“Moral frameworks concerning research on human subjects currently seem to

constitute one region where there exist widely shared understandings of moral

practices and policies.” On the one hand, this statement is accurate. On the other

hand, a bounty of literature in bioethics, empirical and normative, calls this

judgment into question, depending on which parts of these moral frameworks

are under consideration. There currently exists international agreement on major

generalizations such as on most of the vague principles in the Declaration of

Helsinki, but there also exists a plurality of viewpoints on virtually every major

topic in research on human subjects, including such pillars of the system as the

nature of the obligation to obtain informed consent and the proper way to

conduct review of research by committee review.7

THE PROBLEM OF NORMATIVE JUSTIFICATION

DeGrazia asserts that Childress and I “reduce normative ethics to descriptive

ethics” and that we attempt to justify common morality “in terms of consensus.”

I acknowledge that there were at one time flawed sentences regarding the role of

consensus in Principles of Biomedical Ethics, but Childress and I are clear now that


we advance both “normative and nonnormative claims” in defense of the

commonmorality.8 I will here clarify my own views about the distinction between

the descriptive (nonnormative) and the normative.

First, descriptive ethics and consensus reports are factual; normative ethics and

methods of justification are not. Descriptions are not justifications. Claims about

the existence of the common morality need empirical justification, whereas claims

about the justifiability and adequacy of the common morality require normative

justification.

Second, I am not assuming that all persons in all societies do in fact accept the

norms of the common morality. Unanimity of this sort is not the issue. As noted

previously, many amoral, immoral, and selectively moral persons do not embrace

various demands of the common morality. Some persons are morally weak,

others morally depraved. Morality can be misunderstood, rejected, or overridden

by other values.

Third, it is preposterous to hold that a customary set of norms or a consensus

set of norms is justified by the fact of custom or consensus. More unsatisfactory

still is the idea that norms qualify for inclusion in the common morality merely

because they are rooted in custom or consensus. The proposition that moral

justification derives from custom or consensus is a moral travesty. Any given

society’s customary or consensus position may be a distorted outlook that

functions to block awareness of universally valid requirements. Some societies

are in the grip of leaders who promote religious zealotries or political ideologies

that depart from common-morality standards. Such persons may be deeply

committed to their particular outlook—for example, they may be intent on

converting others to their favored political ideology—but they should not be

said to bemorally committed merely because they are committed to a supremely

valued point of view.9 Fanatics in control of the Taliban in Afghanistan and

Pakistan commit horrendous moral offenses. Their enthusiasm is about some-

thing other than the common morality, which to them gives no reason why they

should constrain their particular zealotry.

Fourth, universal agreement about norms that are suitable for the moral life

explains why there is a common morality, but does not justify the norms.10 What

justifies the norms of the common morality, in the pragmatic theory I have

proposed, is that they are the norms best suited to achieve the objectives of

morality. Fundamental moral norms require for their justification that we state

the objective of the institution of morality. Once the objective has been identified,

a set of standards is justified if and only if it is better for reaching the objective than

any other set of standards. This pragmatic approach to justification is my own

preferred strategy for the justification of general moral norms, but I appreciate

that others may prefer a different justification (e.g., a contractarian one). In the

penultimate section of this essay I will supplement this account of pragmatic


justification of common-morality principles with a coherence theory of justifica-

tion for particular moralities.

Fifth, DeGrazia distinguishes between common morality 1 (widely shared moral

beliefs) and common morality 2 (moral beliefs that would be widely shared if

morally committed persons reached reflective equilibrium). I find this distinc-

tion, as presented, puzzling and unsatisfactory. Common morality 2 will generate

different sets of norms that constitute particular moralities, not the common

morality. “Common morality 2” seems to be a general heading for the way in

which particular moralities develop, or perhaps should develop, from the

common morality. Despite these misgivings, this distinction between common

morality 1 and common morality 2 can be salvaged and reconstructed as follows:

Common morality 1 is the set of universally shared moral beliefs as it now exists

(a descriptive claim), whereas common morality 2 is the set of moral beliefs that

ought to be embraced in the common morality (a normative claim). I will return

to this distinction between what is and what ought to be in the common morality

in the final section of this essay.11

SPECIFICATION: MAKING GENERAL NORMS PRACTICAL

The reason why norms in particular moralities often differ is that the abstract

starting points in the common morality can be coherently developed in a variety

of ways to create practical guidelines and procedures. DeGrazia and I agree that

Henry Richardson’s account of specification presents an important way in which

general norms are made suitably practical. Specification is not a process of

producing general norms such as those in the common morality; it assumes

that valid general norms are available. Specifying the norms with which one

starts, whether those in the common morality or norms previously specified, is

accomplished by narrowing the scope of the norms, not by explaining what the

general norms mean. The scope is narrowed, as Richardson puts it, by “spelling

out where, when, why, how, by what means, to whom, or by whom the action is to

be done or avoided.”12

For example, the norm that we must “respect the autonomy of competent

persons” cannot, unless specified, handle complicated problems of what to say or

demand in clinicalmedicine and research involving human subjects. A definition of

“respect for autonomy” (as, say, “allowing competent persons to exercise their

liberty rights”) might clarify one’s meaning in using the norm, but would not

narrow the general norm or render it more specific and contextually appropriate.

Specification is a different kind of spelling out than analysis of meaning. It adds

content. For example, one specification of “respect the autonomy of competent

persons” could be “respect the autonomy of competent patients when they become


incompetent by following their advance directives.” This specification will work

well in some medical contexts, but will confront limits in others, thus necessitating

further additional specification. Progressive specification can continue indefinitely,

but to qualify as a specification, a transparent connectionmust bemaintained to the

initial norm that gives moral authority to the resulting string of norms.

I come now to a critical matter about the way in which particular moralities are

developed through specification. There is always the possibility of developing

more than one line of specification when confronting practical problems and

moral disagreements. Different persons and groups will offer conflicting specifi-

cations of the same norm. In any given problematic or dilemmatic case, several

competing specifications may be offered by reasonable and fair-minded parties,

all of whom are committed to the commonmorality. We should not hold persons

to a higher standard than that of making judgments conscientiously in light of the

relevant basic and specified norms while attending to appropriate factual evi-

dence. It is to be expected that conscientious and reasonable moral agents will

disagree with other conscientious persons over moral weights and priorities in

circumstances that involve a contingent conflict of norms, and it may be that

multiple particular moralities will be developed.

JUSTIFYING SPECIFICATIONS USING THE METHOD OF COHERENCE

I earlier suggested that it is reasonable to hold that the norms of the common

morality are justified pragmatically. In addition to this form of justification,

I accept a coherentist justification that helps determine when the specified

norms that comprise particular moralities are justified.

A specification is justified if and only if there is good reason to believe that it

will maximize the coherence of the overall set of relevant beliefs. DeGrazia and

I share this view. These beliefs could include empirically justified beliefs, justified

basic moral beliefs, and previously justified specifications. This is a version of so-

called wide reflective equilibrium.13 No matter how wide the pool of beliefs, there

is no reason to expect that the process of rendering norms coherent through the

process of specification will come to an end or be perfected. Particular moralities

are continuous works in progress—a process rather than a finished product.

There is no reason to think that morality can be rendered coherent in only one

system through the process of specification. Many particular moralities are

coherent ways to specify the common morality. Normatively, we can demand

no more than that agents faithfully specify the norms of the common morality

with an attentive eye to overall coherence.

The moral life will always be plagued by forms of conflict and incoherence that

need reduction. Our goal should be a method that helps in a circumstance of


conflict and disagreement, not a panacea. Although DeGrazia conjectures that

Childress and I have an “excessive aversion to disagreement,” his thesis must

confront the accusation commonly brought against us that we allow for too much

conflict and disagreement. My view is that we countenance disagreement (in moral

theory and in moral practice) more readily than many writers in bioethics.

Indeed, it appears that we tolerate disagreement more readily than DeGrazia

himself, judging from his comments regarding “standing up for what’s right.”

Standing up for what a person believes to be right is often admirable, but without

more explanation than DeGrazia offers, such resolve can descend into little more

than an insistent inflexibility that refuses to recognize legitimate disagreement.

DeGrazia speculates that our “excessive aversion to disagreement” leads us to

an “excessive accommodation of existing moral beliefs and/or competing con-

stituencies,” resulting in “reduced coherence and justification.” He offers three

examples. I will consider only his third example, which he seems to find the most

flagrant example because “compromise apparently leads to contradiction.” This

example centers on our support for physician-assisted suicide—thereby, he says,

accommodating “liberal commentators”—and the way we take seriously slippery-

slope arguments in opposition to physician-assisted suicide—thus, he says,

accommodating “conservative participants in this debate.” DeGrazia greets with

skeptical disbelief our assertion that these two points of view can be reconciled.

He offers no argument for his conclusion that our position is incoherent and

contradictory, and he does not reconstruct the arguments that Childress and I

present in our book. To see how the two views can be coherently reconciled, we

need before us the distinction that Childress and I use between the justification of

policies and the justification of acts. Public rules or laws sometimes justifiably

prohibit conduct that is morally justified in individual cases. Two moral questions

about physician-assisted suicide need to be distinguished: (1) Are physicians ever

morally justified in complying with first-party requests that they assist patients in

acts of suicide? (2) Is there an adequate moral basis to justify the legalization of

physician-assisted suicide? Childress and I argue that there are justified acts of

assisting patients in committing suicide. However, once public considerations

external to the private relationship between a physician and a patient are brought

into the picture—such as the implications of legalized physician-assisted suicide for

medical education and medical practice in hospitals and nursing homes—these

considerations may provide good and sufficient moral reasons for disallowing

physicians from engaging in such actions as a matter of public law. We argue that

some policies that legalize physician hastening of death would, under at least some

circumstances, be morally problematic. It may turn out that the worries about

legalization advanced in some slippery-slope arguments have force, but it may also

turn out they have little or no force. There is no incoherence in this position on the

multiple sides to the controversy over physician-assisted suicide.


DeGrazia also maintains that Childress and I have had relatively little to say

about the question, What is coherence? He suggests that coherence is simply “the

holistic embodiment of theoretical virtues, the characteristics we expect of any

good theory.” Coherence seems to reduce for him to conformity to the criteria of

good theories. This issue is too involved to be addressed here, but, as DeGrazia

specifically notes, Childress and I have a defensible set of criteria of good theories,

and therefore, by his account, we have an acceptable core notion of coherence.

Nevertheless, he would find it an underdeveloped notion, and I would agree.

CHANGE AND STABILITY IN THE COMMON MORALITY

I turn, finally, to issues about whether morality changes over time. Moral change

entails that what was not previously morally required (or prohibited) is now

morally required (or prohibited). Particular moralities, customary practices, and

so-called consensus moralities can and do change; they may even change by

reversal of a position. For example, a code of research ethics might at one time

endorse placebo-controlled trials only to reverse itself and condemn such trials at

a later time. When relevant circumstances change or new insight is achieved, such

revisions are warranted.

Change in particular moralities is not, however, at issue. The issue is whether the

common morality can and does change. Is it possible in principle or in practice for

this morality to change? Could it come to be the case that we no longer have to

keep our promises, that we can lie and deceive, or that a vice can become a virtue?

To the extent that we can envisage circumstances in which human society is better

served by substantively changing or abandoning a norm in the common morality

(i.e., we can envision that some alternative norm would better serve to counteract

the tendency for the quality of life to worsen or disintegrate in social relation-

ships), change in the common morality could occur and could be justified. For

example, it is possible, however unlikely, that the obligation to tell the truth could

become severely dangerous to the well-being of our friends and acquaintances,

and we might therefore abandon the rule altogether.

Justification of a change in the common morality either would require that one

or more moral norms remain unchanged in the moral system or would require

that the objectives of morality not change. Without some stability of this sort in the

system, moral change is incomprehensible and would lack justification. The point

is that justification of the new norm will require recourse to some unchanged

norms or goals.14

In principle, all moral norms in the common morality could change over time,

but such change is extraordinarily unlikely in practice. It is difficult to construct

actual or even plausible hypothetical examples of a moral principle in the


common morality that has been valid only for a limited duration. Moreover, I do

not believe that we do or will in the future handle problems of profound social

change by altering norms in the common morality. Instead, we will do what we

have always done: As circumstances change, we will find moral reasons for saying

that there is either a valid exception to a particular obligation or a need to specify it

in a somewhat different way. For example, “Do not kill” is a basic moral rule, but

we have never allowed prohibitions against killing to prevail in all circumstances.

Particular moralities have carefully constructed exceptions in cases of war, self-

defense, criminal punishment, martyrdom, misadventure, and the like. We also

will continue to handle social change through the structure of one or more norms

in the common morality overriding one or more different norms.

However, moral change in the common morality may occur in at least one

important respect. Even if the abstract principles of the common morality do not

change, the scope of individuals to whom the principles are deemed to apply has

changed—a problem of moral status. Consider the norm that we should “Treat all

persons with equal moral consideration.” Imagine that we are able to teach both

language and the norms of the common morality to the great apes. It is possible

under these conditions that we would collectively come to the conclusion that the

great apes, not just human apes, should be included in the category of “persons,”

or some comparable category of moral status, and therefore are owed equal moral

consideration. This would be a momentous change in the scope of individuals

covered by the protections of the common morality. It seems unlikely to occur, as

do other changes of this magnitude, but we can conceive of conditions under

which such change could occur.

It can be argued that the common morality has already been refined in a

conspicuously similar manner by changes in the way slaves, women, people of

color, and persons from many other groups have come to be acknowledged

as owed a deeper level of moral consideration than had been previously recog-

nized. This sort of change constitutes a major—and an actual rather than a

hypothetical—modification in the scope of moral rules. But are the changes

that have taken place in recent centuries truly changes in the common morality?

Changes in the way slaves, women, and people of various ethnicities are regarded

seem to be changes in particular moralities or ethical theories rather than in the

common morality.

The most defensible view, I suggest, is that the commonmorality does not now,

and has never in fact, included a provision of equal moral consideration for all

individuals. However, this norm could become part of the common morality,

which would constitute a substantive modification. I am confident, however,

that empirical investigation of rules determining who should receive equal con-

sideration would show considerable disagreement across individuals and socie-

ties, even among people who could reasonably be said to be committed to the


objectives of morality. This finding is consistent with the normative thesis that the

common morality should include rules of equal moral consideration for slaves,

women, people of color, and other relevant parties now excluded.

Descriptive ethics analyzes where we are (which rules we do accept); normative

ethics ventures into the waters of where we should be (which rules we ought to

accept).Wherewe are in the commonmorality is not necessarilywherewe should be

according to a normative theory. By appeal to what I have called the objectives of

morality, we can (arguably) justify the claim that rules of equal moral consideration

ought to be applied to all persons—not merely to limited groups of persons. I will

not attempt to justify this normative thesis, but I wish to mark its importance.

Among themost momentous changes to occur in the history ofmoral practice have

been those regarding the scope of persons to whom moral norms are applied. A

theory of the common morality that deprives it of the capacity to criticize and

evaluate existing groups or communities whose viewpoints are morally deficient

would be an ineffectual theory. A profoundly important feature of the common

morality is its provision of cross-cultural standards of evaluation. To the extent that

the common morality itself stands in need of improvement, we can hope to make

those improvements by revising the normative guidelines necessary to achieve the

fundamental objectives of morality. We need to think not merely in terms of what

universally is the case, but also in terms of where we might be in the future.

Notes

1. Although there is only one universal commonmorality, there is more than one theory

of the common morality. For a diverse set of theories of the common morality, see Alan

Donagan, The Theory of Morality (Chicago: University of Chicago Press, 1977); Bernard

Gert, Charles M. Culver, and K. Danner Clouser, Bioethics: A Return to Fundamentals (New

York: Oxford University Press, 1997); and W. D. Ross, The Foundations of Ethics (Oxford:

Oxford University Press, 1939).

2. See Sissela Bok, Common Values (Columbia, MO: University of Missouri Press, 1995),

13–23, 50–59 (citing several influential writers on the subject).

3. Compare the arguments in G. J. Warnock, The Object of Morality (London: Methuen

& Co., 1971), esp. 15–26; JohnMackie, Ethics: Inventing Right andWrong (London: Penguin,

1977), 107ff. I have been deeply influenced on this subject by Warnock and Mackie, and no

less by Thomas Hobbes and David Hume.

4. See Richard B. Brandt, “Morality and Its Critics,” in hisMorality, Utilitarianism, and

Rights (Cambridge: Cambridge University Press, 1992), chap. 5.

5. I concur that virtually all of the norms mentioned by Turner are not in the common

morality and that we can differentiate societies by the different norms that each society

accepts.

6. When I say that some persons are not committed to morality, I do not mean that they

are not dedicated to a way of life that they consider a moral way of life. Religious fanatics


and political zealots clearly have this self-conception even as they act against or are

neglectful of the demands of morality.

7. For an overview of persistent problems and controversies, see Committee on

Assessing the System for Protecting Human Research Participants, Institute of Medicine,

Responsible Research: A Systems Approach to Protecting Research Participants (Washington,

DC: The National Academies Press, 2002).

8. Tom L. Beauchamp and James F. Childress, Principles of Biomedical Ethics. 5th edn.

(New York: Oxford University Press, 2001), 4.

9. For a once influential literature suggesting that moral norms are those that a person

or, alternatively, a society accepts as supreme, final, or overriding, see The Definition of

Morality, ed. G. Wallace and A. D. M. Walker (London: Methuen, 1970), and William

K. Frankena, Perspectives on Morality, ed. K. E. Goodpaster (Notre Dame, IN: University of

Notre Dame Press, 1976), chaps. 10, 15. A criterion of supremacy would permit almost

anything to count as moral if a person or a society is committed to its overriding pursuit.

10. Cf. Mackie, Ethics: Inventing Right and Wrong, 22–23.

11. I agree with DeGrazia that there are many forms of “moral prejudice” and that

“moral judgment can be distorted in many ways.” However, I disagree with him that these

distorted judgments “reflect commonmorality 1.” These judgments do not reflect common

morality as I use the term.

12. Henry Richardson, “Specifying, Balancing, and Interpreting Bioethical Principles,”

Journal of Medicine and Philosophy 25 (2000): 285–307, quote on p. 289. See also

Richardson, “Specifying Norms as a Way to Resolve Concrete Ethical Problems,”

Philosophy and Public Affairs 19 (1990): 279–310; David DeGrazia, “Moving Forward in

Bioethical Theory: Theories, Cases, and Specified Principlism,” Journal of Medicine and

Philosophy 17 (1992): 511–539; DeGrazia and Beauchamp, “Philosophical Foundations and

Philosophical Methods,” in Daniel Sulmasy and Jeremy Sugarman, eds. Methods in

Medical Ethics (Washington, DC: Georgetown University Press, 2001), 31–46.

13. Norman Daniels, “Wide Reflective Equilibrium and Theory Acceptance in Ethics,”

Journal of Philosophy 76 (1979): 256–282; Daniels, “Wide Reflective Equilibrium in

Practice,” in L. W. Sumner and J. Boyle, eds., Philosophical Perspectives on Bioethics

(Toronto: University of Toronto Press, 1996).

14. See the argument to this conclusion in Joseph Raz, “Moral Change and Social

Relativism,” in Ellen Paul, Fred Miller, and Jeffrey Paul, eds., Cultural Pluralism and

Moral Knowledge (Cambridge: Cambridge University Press, 1994), 139–158.


11

FROM MORALITY TO COMMON MORALITY

For some 19 years, Bernard Gert has feasted on my views about moral

philosophy and the principles of biomedical ethics. Between 1990 and 1994,

Gert and his coauthors published a group of articles1 critical of my work

with James Childress.2 These articles were widely read in biomedical ethics,

and Gert has since continued to engage my work. Rather than answering his

criticisms, I here consider a single problem at the heart of Gert’s moral

philosophy: the justification of claims about the common morality. My goal

is to identify and provide a limited defense of three forms of justification of

claims about the common morality.

Despite Gert’s many efforts, and my own, the common morality is little

studied in contemporary philosophy and rarely embraced as foundational in

a moral theory. In much of the literature on the subject, Gert and I, together

with our coauthors, are treated as the only defenders of common morality

theory. This assessment may be correct. Clearly many philosophers are

skeptical of appeals to a common morality. My graduate students, my

colleagues, and the critics3 Gert and I share in common have frequently

recommended that I abandon common morality theory. Their advice is that

I rely exclusively on considered moral judgments and reflective equilibrium,

following the path charted by John Rawls, Norman Daniels, and others.4

However, both Gert and I have rejected this advice.5

A HISTORY OF THE GERT CORPUS: HOW COMMON MORALITY

BECAME DOMINANT

I start with a brief history of Gert’s movement from morality to the common

morality. His first published book was entitled The Moral Rules: A New Rational

Foundation for Morality.6 This title draws attention to three notions: “the moral

rules,” “morality,” and “rational foundation.” Gert was at the time centering his

theory on rules, rationality, and morality. Eighteen years later, in a 1988 revision,

the term rational was removed from the subtitle and the revised book was issued

as Morality: A New Justification of the Moral Rules.7 This “new justification” in

1988 was fundamentally identical to the justification in the 1970 book. Another

revised edition in 1998 removed the word new from the subtitle.8 As this history of

titles indicates, over a period of 35 years of publication (now 40 years ago) only

one word—morality—has remained constant in the titles of Gert’s theoretical

masterwork.

In 2004, Gert published his Common Morality: Deciding What to Do.9 This title

suggests that the term morality may be on the way to replacement by the term

common morality. The first appearance of the term common morality in the

publications of the Gert group10 (or the “Dartmouth descriptivists,” as they

once described themselves) known to me was published in 1995 by Gert’s friend

and longtime coauthor, Dan Clouser, in an article he published specifically as a

criticism of my work with Childress.11 In the same year Sissela Bok published her

underappreciated book Common Values.12 Childress and I had begun to use the

term common morality in our edition published in 1994, under the influence of a

1977 book by Alan Donagan that used the language of “common morality.”13

Something seems to have been in the air around 1994 that led to the wider use of

the term “common morality.”14

THE NATURE OF THE COMMON MORALITY

Gert and I both understand the common morality as universal morality.15 It is not

relative to cultures or individuals and is to be distinguished from norms that bind

only members of special groups such as cultures, religions, or professional associa-

tions. All persons committed to morality accept its core moral norms. They know

not to lie, not to steal others’ property, not to kill or incapacitate innocent persons,

to keep promises, to respect the rights and liberties of others, and the like.

Gert views the nature and number of the rules in the common morality as

precisely determinable,16 whereas I regard these matters as less than precisely

determinable. I will not address this issue, but I note that it is of the highest

importance in the interpretation of Gert’s writings that we not concentrate


excessively on the moral rules. “Morality” references the entire “moral system” for

him. His full system includes moral rules, moral ideals, and a two-step procedure

for determining when a violation is justified. An undue emphasis on rules can

cause neglect of the full system, which in turn can cause us to misunderstand how

Gert views the common morality and to interpret him as having an unduly

narrow moral theory. When he says that the “phrase ‘moral system’ . . . has the

same meaning as ‘morality,’”17 he means that “morality” includes the full spread

of shared moral norms, whether or not they are norms of moral obligation.

THE JUSTIFICATION OF CLAIMS ABOUT THE COMMON MORALITY

I come now to my main question: Which types of justification of the common

morality may be offered, and for which types of claims about the common

morality are they suitable?

I distinguish three types, or strategies, of justification: (1) normative theoretical

justification of the sort found in ethical theories; (2) normative conceptual

justification that rests on conceptual analysis; and (3) empirical justification

reached through empirical research. I do not claim that each each of these

strategies of justification justifies the same conclusion or set of conclusions

about the common morality, nor do I claim to produce here a justification that

uses any of the three strategies. Rather, my limited aim is to identify three available

types of justification. In bringing empirical justification into this picture, I

obviously must distinguish justification of norms of the common morality

(Gert’s principal project) from justification of claims that there is a universally

shared common morality (which is, as yet, of little or no interest to Gert).18

NORMATIVE THEORETICAL JUSTIFICATION

Gert has argued convincingly that the norms of the commonmorality can be justified

by a moral theory. This is not to say that he has convinced me of the rightness of his

moral theory in particular. Imean that he has shown that an ethical theory can be put

to the work of justifying the norms of the commonmorality. In his theory common

morality is justified on the basis of rationality. To justifymorality, onemust show that

it is a public system that all rational persons would accept as universally valid. Gert

regards it as clear to all rational persons that we should not act irrationally because

irrational actions are those that should not be performed:

Rational persons want to avoid death, pain, disability, loss of freedom,

and loss of pleasure, and they know not only that they are fallible and

F R OM MO R A L I T Y T O C OMM O N MO R A L I T Y 191

vulnerable but that they can be deceived and harmed by other people.

They know that if people do not act morally with regard to them, they will

be at significantly increased risk of suffering some harm. If they use only

rationally required beliefs, it would be irrational not to endorse common

morality as the system to be adopted to govern the behavior of all moral

agents.19

Acting irrationally has a close relationship to acting in ways that will increase

the likelihood of certain basic harms, and Gert argues that moral rules have the

goal of prohibiting the causing of these harms or contributing to conditions that

may cause them.20

Ethical theories other than Gert’s could be employed to justify the common

morality. Gert himself often suggests that Hobbes and Kant more or less had this

strategy in mind,21 and he has commented that, “Like Mill, most moral philoso-

phers begin by trying to see what support, if any, can be given to common

morality.”22 I have previously used a form of pragmatism to illustrate a type of

theory that is particularly suited to serve this purpose; this theory also has

similarities to Gert’s conception of the goals of morality.23 Pragmatic justification

holds that moral norms are justified by their effectiveness in achieving the object

of morality. In its most general formulation, pragmatic justification can be used to

justify a wide array of types of norms, not merelymoral norms: Once an operative

purpose or objective of an institution or system of thought has been identified, a

set of standards is consideredmost suitable, and therefore is vindicated, if it can be

shown to be better for reaching the identified objectives than any alternative set of

standards.24 I am not altogether endorsing pragmatic justification. I am main-

taining that pragmatic justification is positioned to provide a theoretical justifica-

tion of the common morality that has similarities to Gert’s recent statements

about justification.

Notably close to Gert’s formulation of the goal of morality is the pragmatist

starting point that I call, using G. J. Warnock’s term, “the object of morality.” Gert

likely would regard such pragmatic justification as threatening the secure status of

the privileged norms of common morality and as introducing too much con-

tingency into the system of basic norms and ideals of morality. He would be

concerned that the theory would justify a set of norms other than those of the

common morality, much as utilitarianism does. Nonetheless, Gert’s description

of the goal of morality is conspicuously similar to the starting point that I am

proposing for pragmatic justification. Gert has consistently argued that morality’s

goal is the lessening of evil or harm and has insisted that this goal is a major

characteristic of morality.25While only slightly modifying this account of the goal

of morality, a pragmatist can conceive the goal as that of promoting human

flourishing by counteracting conditions that cause the quality of people’s lives to


worsen. The goal is to prevent or limit problems of indifference, conflict, suf-

fering, hostility, scarce resources, limited information, and the like. We have

learned from centuries of experience that the human condition tends to deterio-

rate into misery, confusion, violence, and distrust unless norms of the common

morality are enforced through what Gert calls an informal public system—that is,

a public normative system which, by not having authoritative judges and formal

decision procedures, is structured informally.26 When complied with, these

norms lessen human misery, foster cooperation, and promote human flourishing.

Such norms may not be necessary for the survival of a society, as some have

maintained,27 but it is reasonable to claim that they are necessary to ameliorate the

tendencies for the quality of people’s lives to worsen and for the disintegration of

social relationships. This formulation of the object of morality bears a closer

resemblance to Warnock’s theory than to Gert’s,28 but Gert’s formulations are

similar to my formulations, and we are both indebted to elements of Hobbes’s

theory, especially his accounts of human nature and the human condition.

In the pragmatist theory as I conceive it, the norms of morality are those the

evidence indicates best serve the object of morality. That is, once the object of

morality has been identified, a set of standards is pragmatically justified if and

only if it is the best means to the end identified when all factors—including

human limitations, shortcomings, and vulnerabilities—are taken into considera-

tion. If one set of norms will better serve the object of morality than a set currently

in place, then the former should displace the latter. It is here that Gert will see

pragmatic justification as giving only contingent support to moral rules. I will not

now pursue this problem, but later I will discuss the contingency of moral norms

in general and whether it is plausible to believe that there are good candidates to

displace what Gert and I consider to be the norms of the common morality.

In other publications I have supported the method of reflective equilibrium as a

valid form of justification. It might seem inconsistent that I do not call on it in the

present discussion of justification in ethics. However, reflective equilibrium

theory is not a candidate for justifying the common morality. Reflective equilib-

rium needs the commonmorality to get off the ground, and so is in no position to

offer a justification of it. It would obviously be circular to use the norms of the

common morality to justify the norms of the common morality. Rawls says that

reflective equilibrium begins with considered judgments, none of which is foun-

dational and all of which are provisionally fixed points, subject to emendation or

even elimination. Rawls uses this term to refer to “judgments in which our moral

capacities are most likely to be displayed without distortion.”29My view is that the

norms of the common morality are considered judgments and are most unlikely

to be subject to alteration by reflective equilibrium.30

Another problem of circularity also deserves a brief comment. It might be argued

that ethical theories cannot provide a noncircular justification of the common


morality because they presuppose the same norms the theory is presumed to justify,

just as reflective equilibriumdoes.Here the question is, “Howcan a theory ofmorality

proceed without presuming morality?” However, this objection is not relevant to

what Gert and I have proposed. Gert’s theory of rational justification does not

presuppose particular norms of the common morality, nor does the pragmatic

justification that I have sketched. In pragmatic justification, many norms could

qualify as justified norms of morality, and it could turn out that there is no single

justified morality. That is, there might be no common morality at all.

NORMATIVE CONCEPTUAL JUSTIFICATION

Now I come to a thornier issue: Can the norms of the commonmorality be justified

by analysis of the concept ofmorality? The problem originates in the looseness of the

ordinary-language use of the termmorality and a lack of clarity in the way philoso-

phers commonly employ the term. To shed light on themeaning of “morality,” Gert

provides an instructive distinction, which I will here adopt. He distinguishes

between the descriptive and the normative senses of the term. The descriptive sense

has no implications for how persons should behave, whereas the normative sense

determines that some behaviors are immoral and others morally required.31

Morality in the Descriptive Sense

In the descriptive sense, “morality” refers to a group’s code of conduct, or perhaps to

important beliefs and attitudes of individuals. In this sense moralities can differ exten-

sively in the content of beliefs and in practice standards. One society might emphasize

the liberty of individuals, while another emphasizes the sanctity of human and animal

life. Rituals may be established in one society, but disavowed in another. What is

immoral inone societymaybe condoned, or evendeemedmorally required, in another.

“Morality” in the descriptive sense needs no further examination here, because it

will not help us in identifying types of justification of the norms of the common

morality. However, I will return inmy final section to ways in which social scientists

interested in descriptive morality can constructively investigate whether a common

morality exists and can provide empirical justification of their findings.

Morality in the Normative Sense

Gert writes that, “‘Natural law’ theories of morality . . . claim that morality applies

to all . . . persons, not only those now living, but also those who lived in the past.

These are not empirical claims about morality; they are claims about what is


essential to morality, or about what is meant by ‘morality’ when it is used

normatively.”32 These claims are conceptual. Gert seeks the substantive norms

that constitute the common morality and are essential to the concept of morality.

He sometimes presents the questions as definitional, rather than conceptual and

justificatory, but I will not address these distinctions.

Moral pluralists claim that there are multiple concepts of morality in the

normative sense, but moral pluralism is a group-relative notion best interpreted

as a version of descriptive morality. It would be incoherent to formulate the

normative meaning of the term morality as composed of the norms of multiple

moralities, because contradictory advice would be given. One could deny that the

term is univocal and then formulate two or more normative senses (ns1, ns2, etc.),

each with a different set of substantive norms, just as we can distinguish descrip-

tive and normative senses. However, this maneuver is again the functional

equivalent of analyzing morality descriptively rather than normatively.

Accordingly, I will not further pursue conceptual pluralism.

Gert’s natural law approach enables him to connect his views, as he should, to

traditional theories that take a universal normative approach. His personal

inspiration is Hobbes, but other philosophers serve just as well. For example,

Locke seems a perfect fit for Gert when he writes that “the law of nature [is] plain

and intelligible to all rational creatures, yet men [when] biassed by their interest,

as well as ignorant for want of studying it, are not apt to allow of it as a law binding

to them in . . . their particular cases.”33

Extensive discussion of Gert’s views about the essential normative content of

morality is beyond the scope of this paper, but a brief observation will serve to

situate his account as conceptual in nature. In the 2005 revised edition of his

masterwork, Gert has brought part-and-chapter divisions to his book for the first

time in 35 years. Part I, which contains six chapters, is titled “Conceptual

Foundations.” This part, when joined with his Stanford Encyclopedia article,

shows that he conceives his work as conceptual: The project is a conceptual

analysis of “morality” in the normative sense. Part II, “The Moral System and

Its Justification,” provides Gert’s justification of the common morality and is an

example of what I earlier discussed as normative theoretical justification. It might

seem that Gert here presents a moral justification of his normative theory, but this

is not his stated goal. He says that he offers a justification of a public system of

morality that applies to all moral agents.34

Gert’s work in the early chapters is a wide-ranging conceptual analysis of

morality in the normative sense, but his work can be misunderstood because of

its movement between the language of the conceptual analysis of morality and that

of the nature of morality. There also may be unclarity as the language shifts

between “morality,” the “concept of morality,” and a “theory of morality.”

However, I will here be satisfied to agree with Gert that there is a concept of

F R O M MO R A L I T Y T O C OMM O N MO R A L I T Y 195

“morality” in the normative sense and that its analysis requires the provision of

specific normative content in the form of rules and ideals.

Nonetheless, I worry that what Gert claims to discover in the concept of morality

reduces it, more than he acknowledges, to the normative content of his ethical

theory and to certain presumptions he makes about the nature of the common

morality. He denies it, however, and adds the following perceptive observation:

The differences in content among those philosophers who use “morality” to

refer to a universal guide that all rational persons would put forward for

governing the behavior of all moral agents are less significant than their

similarities. For all of these philosophers, such as Kurt Baier, Philippa Foot,

and GeoffreyWarnock, morality prohibits actions such as killing, causing pain,

deceiving, and breaking promises. . . . As Hobbes said in both Leviathan andDe

Civemorality is concerned with promoting people living together in peace and

harmony, not causing harm to others, and helping them. . . . The differences

among those philosophers who hold that there is a universal morality are

primarily about the foundation of morality, not about its content.35

This passage correctly acknowledges that many philosophers with different

conceptions of the theoretical justification of universal morality largely agree on

the general norms that comprise morality in the normative sense, even though

they may disagree on justified exceptions to those norms. Put another way,

philosophers converge on the principles, rights, and responsibilities that are

essential to morality, and in doing so converge on essential conditions of the

concept of morality.36 They also agree on paradigm cases such as the judgment

that rules legitimizing the slave trade are unacceptable independent of any

cultural framing of ethics and independent of any framework of law, whether

past or present, national or international. Slavery is inconsistent with morality,

and rights against being treated as a slave are cross-culturally valid. As Locke’s

comment implies, it is plain to any reasonable person who turns attention to the

notion of morals that slavery is morally wrong and inconsistent with “morality” in

the normative sense. I understand Gert to stand in this tradition of moral

philosophy, and I locate myself there as well.

The Definition of Morality

How, then, shall we define the term morality in the normative sense? Gert

proposes the following definition: “Morality is an informal public system

applying to all rational persons, governing behavior that affects others, and has

the lessening of evil or harm as its goal.”37 This definition is acceptable for the


most part, but I prefer to modify it slightly. Instead of “all rational persons,” I

prefer “all persons capable of moral agency” (a formulation Gert himself some-

times uses38); and instead of “lessening of,” I prefer to be specific about whether

“lessening of” is restricted to “not causing” harm or is a broader notion. Gert

acknowledges that his notion of “lessening” (which is not the term he always uses

to express his views on this subject39) is controversial: “The final characteristic of

morality—that it has the lessening of evil or harm as its goal—is also somewhat

controversial. . . . [But] even those precepts that require or encourage positive

action, such as helping the needy, are almost always related to preventing or

relieving harms.”40

This formulation in terms of preventing or relieving is an important clarifica-

tion of his views, as is the previously quoted description Gert offers when

mentioning Hobbes (“morality is concerned with promoting people living

together in peace and harmony, not causing harm to others, and helping

them”). Gert seems to have eased up on his nonmaleficence-oriented conception

of the moral rules at the heart of the content of “morality” in the normative sense,

though his theory has never been narrowly nonmaleficence-oriented in the con-

tent of the moral ideals. In his article on the “Definition ofMorality” and in recent

book revisions, he implies, however guardedly, that we can set aside disputes

about whether nonmaleficence shades into beneficence.41Gert is, in my judgment,

now well positioned to work out any looming inconsistency or narrowness in his

earlier work by using his distinction between the moral ideals and the moral rules.

Gert and I are agreed that the concept of morality will not be analyzed appro-

priately unless there is a delineation of the particular set of substantive norms that

are essential to morality. Philosophers who attempt to analyze the concept of

morality exhaustively in terms of formal or nonsubstantive conditions will miss

what is most important. Examples are theories that analyze morality in terms of

(1) norms that are regarded as supremely authoritative and of overriding social

importance; (2) norms that are prescriptive in form (i.e., action-guiding impera-

tives that do not describe states of affairs); or (3) norms that harmonize pro and

con interests.42 These once popular approaches to the definition of morality—

possibly the received approach in moral philosophy—are ways of investigating

how conceptually, we can identify something as a morality. They are appropriate

and useful in the context of the study of morality in the descriptive sense; but, by

design, these accounts do not address whether there is a specific normative

content that is privileged and constitutive of morality.

My work, and Gert’s, would probably be improved by making it clear that the

conceptually privileged content of morality in the normative sense entails that

certain moral norms that are discovered through work in descriptive ethics are

conceptual errors and are necessarily false (which is not to say they are absolute

rather than prima facie). For example, it is a conceptual mistake, when using


morality in the normative sense, to assert that morality allows persons to trade in

slaves. The proposition that the slave trade is permissible might be true of the

morality of certain groups when “morality” is used in the descriptive sense, but it

is conceptually incorrect when speaking of morality in the normative sense.

Similarly, the proposition that “lying is morally permissible” is necessarily false;

it contradicts the norm “lying is not morally permissible,” which is a conceptually

privileged norm in “morality” in the normative sense. “Lying is morally permis-

sible” is not necessarily false when using “morality” in the descriptive sense, but it

is necessarily false in the normative sense.

Similarly, it is inconceivable that a person who rejects the principle of nonma-

leficence, or who makes vices into virtues and virtues into vices, has coherent moral

beliefs in the normative sense of morality, even if he or she has a self-perception of

embracing morality. Morality in the normative sense has here been stretched

beyond its conceptual boundaries. Similarly, its boundaries are exceeded if one

asserts that a good mariner is necessarily a morally good person or that all judg-

ments about good tennis are moral judgments. Some sort of conceptual confusion

is at work. However, I will not further address ideas about the methods of analysis

that might be used to investigate the concept of morality.

I conclude this section by returning to the definition of morality in the

normative sense. As noted earlier, Gert states, as a definition, that “Morality is

an informal public system applying to all rational persons, governing behavior

that affects others, and has the lessening of evil or harm as its goal.” I have

proposed modifications, and, accordingly, recommend the following formula-

tion, which follows Gert’s wording as far as my own analysis allows: Morality in

the normative sense is a single informal public system of norms with authority to

guide the actions of all persons capable of moral agency in their behavior that

affects others; its goal is to prevent or limit problems of indifference, conflict,

hostility, vulnerability, scarce resources, and the like. This definition is reasonably

consistent with Gert’s recent formulations and explanations using the language of

lessening.43

This definition does not provide a full conceptual analysis of morality in the

normative sense. I have not provided such an analysis, whereas Gert has

attempted to do so. Whether he has done so successfully will depend on a close

assessment of his work in Part I of Morality: Its Nature and Justification.

Conceptual Change

My suggestions so far leave unresolved whether it is possible for conceptually

privileged norms of morality to change, in which case it could not be claimed that

there is a conceptually essential body of norms constituting morality in the


normative sense. If norms are not fixed, then they could evolve into different

norms, thereby altering the concept of morality in the normative sense. In Gert’s

theory, change cannot occur in the norms of morality because the basic moral

rules are essential and not time bound: “A general moral rule concerns actions

open to all rational persons in all societies at all times. . . . A general moral rule is

unchanging and unchangeable; discovered rather than invented. . . . Since general

moral rules apply to all rational persons at all times, obviously they cannot be

invented, or changed, or subject to the will of anyone.”44 But is Gert correct in this

assessment?

To the extent that we can envisage circumstances in which human society is

better served by substantively modifying or abandoning a norm in the common

morality, change could in principle occur and could be justified under the account

of pragmatic justification that I earlier sketched. It is conceivable, for example,

that the rule that we are obligated to tell the truth could become sufficiently

dangerous that we would abandon the rule of truth telling altogether. However

unlikely, the possibility of such change might be taken to weaken the claim that

there is a common morality with essential conditions and that holds a timeless

normative authority for all moral agents.

It is easy to see that morality in the descriptive sense has changed and continues

to change, but it is not clear whether such change actually might occur in morality

in the normative sense. It would be dogmatic to assert without argument that the

privileged norms of “morality” in the normative sense cannot change, but it is

difficult to construct a historical example of a basic moral norm that has been

valid for a limited duration and was dropped because some good moral reason

was found for its displacement. No evidence known to me suggests that societies

have handled moral problems by either rejecting or altering basic norms in the

common morality. As circumstances change, we find moral reasons for saying

that there are valid exceptions to a norm or new specifications of it, but we do not

discard basic norms.

Nonetheless, I do not conclude, as Gert does, that morality in the normative

sense is a web of norms so fixed that those norms cannot be altered, no matter the

circumstances. It is not difficult to envision a situation in which a rule might be

added to the common morality, by contrast to rules being abandoned or swapped

out. For example, the commonmorality could be expanded to include a basic rule

of equal moral consideration (or nondiscrimination), and inclusion of such a rule

could constitute a substantial change in the common morality.45 I am assuming

that a rule of equal moral consideration is not a basic rule in the common

morality. There lurks here the problem that rules of nondiscrimination are similar

to antislavery rules: Historically neither has been socially sanctioned in most

societies for everyone in those societies. Slavery, like other forms of discrimina-

tion, and in every society in which slavery abounds, has always involved scope


questions about which group is protected by the rules of morality. I do not believe

that the common morality has ever addressed the question of precisely who is

protected. However, this issue about scope questions would distract from the

current argument and I put it aside here.46

My ideas about the pragmatic justification of moral norms clearly suggest that,

were human nature or human social conditions to be altered sufficiently, a change

of moral rules to accommodate these changes might be justified. Gert rejects this

possibility, but perhaps without a full consideration of the range of possible

changes that might occur. It should be apparent by now that I find Gert’s thesis

that the norms of morality are atemporally fixed to be an overly demanding

conceptual condition of morality in the normative sense. A less demanding

condition, and one I think more plausible, is that there is a body of basic moral

norms that distinguish universal morality, or the common morality, from parti-

cular moralities. Particular moralities certainly change, but even universal mor-

ality allows for the possibility of conceptual change. At the same time, it is

empirically unlikely that there will be shifts in universal morality, and for this

reason it is unlikely that the normative concept of morality will undergo change.

Any philosopher’s position on these questions is sure to be influenced by his or

her more general theoretical commitments. My defense of the possibility of moral

change is facilitated by the fact that I appeal to pragmatic justification: If contexts

shift sufficiently, new norms may be required to improve protections from harm.

Gert is not committed to pragmatic justification or to social contingencies as

important considerations, and his commitments take him down a different path.

From this perspective I cannot be said to have shown that his views are not

correct, and such an argument is more than I will undertake here.

EMPIRICAL JUSTIFICATION

I move on now to the third type, or strategy, of justification. Some have opined

that, as Gert and I conceive it, “commonmorality theory is empirical in nature.”47

But are questions about the commonmorality empirical matters? And whatmight

it mean to say that there could be an empirical justification of the claim that a

common morality exists?

As far as I can determine, no empirical studies throw into question whether

some cultural moralities accept, whereas others reject, what Gert and I describe as

the norms of the common morality. Existing empirical data are, of course, from

studies of moralities in the descriptive sense. These empirical investigations have

studied differences in the way rules of the common morality have become

embedded and applied in cultures and organizations, but they more assume

than investigate what Gert and I have in mind when we speak of the common


morality. Empirical studies succeed in showing cultural differences in the inter-

pretation and specification of moral norms, but they do not show whether

cultures accept, ignore, abandon, or reject the standards of the commonmorality.

For example, empirical studies have not tested whether certain cultural moralities

reject rules against theft, promise breaking, or killing. Instead, investigators study

the conditions under which these acts are judged to occur in cultures, how

cultures handle exceptive cases, and the like.

Nonetheless, facts about moralities in cultures and the morality held (or not

held) in common are important for philosophers interested in the common

morality, and they are more important for Gert’s claims about the common

morality than he acknowledges. He says that “Every person subject to moral

judgment knows what kinds of actions morality prohibits, requires, discourages,

encourages, and allows.”48 Gert also writes that the common morality exists

independent of a moral theory,49 that “the existence of a common morality is

supported by widespread agreement on most moral matters by all moral

agents,”50 and that “I regard morality or the moral system as the system people

use, often unconsciously, when they are trying to make a morally acceptable

choice.”51 He argues that although societies regard morality as concerned pri-

marily with practices that prevent or minimize harms, it is “doubtful that any

actual society has a morality that contains only this feature. . . . Nonetheless, this

feature of morality . . . is included in everything that is regarded as morality in all

societies.”52 And to all of this Gert adds, “There are many religions but only one

morality.”53

These claims are empirical, not meta-ethical or normative. Gert seems to claim

(1) that all rational persons know the same things about basic moral conduct, (2)

that all societies have a morality oriented around the feature of avoiding and

minimizing harm, and (3) that it is doubtful whether avoiding and minimizing

harm are included in the full scope of what is regarded as morality in any society.

Empirical investigation is needed to support these statements.54 Otherwise, they

are merely speculative generalizations drawn from anecdote, limited experience,

and philosophical theories of human nature.

Despite his own apparent reliance on empirical claims, Gert has been

unfriendly to the idea that important claims about the common morality rest

on factual assumptions that merit empirical study. His view seems to be one of a

distanced disregard of the relevance of empirical evidence. His work presumably

relies exclusively on normative theoretical content and normative conceptual

analysis, as informed by a theory of human nature. He seems to think that no

additional form of inquiry bears on either his claims or his assumptions. Gert says:

It is very tempting to think that I am simply starting with an account of

human nature, including an account of human rationality, fallibility, and


vulnerability, and simply generating a code of conduct that all rational

persons would favor adopting to guide everyone’s behavior. But I am not

doing this, because it would not count as justifying morality unless the code

of conduct being justified was virtually identical to the moral system that is

now implicitly used in deciding how to act morally and in making moral

judgments. . . . The starting point has to be common morality.55

I do not disagree with this statement, but I would add to it that we are certain to

encounter different conceptions of how to formulate the precise content of the

common morality, including the content of what Gert calls the starting point.

This formulation is not a purely philosophical matter, and I believe that empirical

work could be conducted that would add additional support to a general program

defending the claim that there exists a common morality of a certain sort, by

contrast to the claim that there is an adequate justification of the common

morality. Such empirical work has the potential to help clarify claims of the

actual moral content in the common morality. Scientific studies sometimes

confirm our beliefs, but at other times they upset or correct our beliefs. As

things now stand, we can only create hypotheses about the proper starting point.

Oliver Rauprich has argued, I think appropriately, that common morality

theory, if it is to be credible, must more carefully address empirical claims

about the status of a shared morality. He proposes that studies should be designed

to capture the moral beliefs that prevail in societies and that are shared by a vast

majority of people. He recommends that investigators study the extent to which

there is shared agreement. For example, he thinks we need to find out whether

“shared beliefs” means that, say, “95 or even 99%of the people” share the norms.56

A frequent criticism of common morality theory is that anthropological and

historical evidence already speak against the empirical hypothesis that a universal

morality exists.57 I earlier suggested that this thesis misconstrues the findings of

historians and anthropologists. I add now that these critics of common morality

theories, unlike Rauprich, seem to me unappreciative of the nuances that would

surround the design of empirical research on the common morality. In principle,

scientific research could either confirm or falsify the hypothesis that there exists a

universal commonmorality containing the norms that Gert and I envisage, but, as

with all empirical research, it must be stated which hypotheses are to be tested,

how the inclusion/exclusion criteria for study subjects are to be formulated, and

why these hypotheses and criteria were selected. To date, critics who appeal to

available empirical studies as falsifying common morality claims have not

attended to the specific concerns that must be addressed in the scientific investi-

gation of hypotheses about the common morality.

What are these concerns? I have elsewhere emphasized the importance of an

inclusion criterion for study subjects that would allow only persons who are


committed to morality.58 Persons not committed to morality, such as criminals

and sociopaths, should be excluded because their inclusion introduces a fatal bias.

When I say that some persons are not committed to morality, I do not mean that

they are not dedicated to a way of life that they consider a moral way of life.

However, this commitment may be no more than a set of intolerant and narrow-

minded beliefs—a commitment of little interest in confirming or falsifying

hypotheses about the common morality.

It would be devious of me to deny that many millions of people from many

societies likely would be excluded by this criterion, which leads to its own nest of

theoretical and practical problems. Here I think both philosophers and social

scientists have to bite the bullet and assert that massive swaths of some societies

are indeed inappropriate subjects because of their beliefs. It should not matter

whether the number of persons excluded is large or small: The issue is whether

inadequate impartiality or concern for others renders persons inadequate as test

subjects, and I think that the answer is yes. Even if these individuals are committed

religious persons of emulable moral integrity within the framework of their

beliefs—for example, certain Christian, Jewish, or Muslim fundamentalists—

they may not have a proper impartiality and concern for the welfare of others

beyond their affiliations.

The main hypothesis that I propose be empirically tested is this: All persons

committed to morality and to impartial moral judgment in their moral assessments

adhere to the norms in what Gert and I have called the common morality. The

persons to be included in the study, with other potential subjects excluded, are (1)

persons who pass a rigorous test of whether their beliefs conform to some critical

component feature of the normative sense of morality (with this feature to be

designated in the study design); and (2) persons who qualify as having the ability

to take an impartial moral point of view.

It would be difficult, but not impossible, to design this study without either

missing the target—namely, the beliefs of all and only those who are committed to

both morality and impartiality—or begging the question by insisting on studying

only persons known to accept what Gert and I call morality in the normative

sense. The question could be begged either by (1) designing the study so that the

only persons who are included as subjects are those who already have precisely the

commitments and beliefs the investigator is testing for or (2) designing the study

so that all persons are tested whether or not they are committed to morality and

impartiality. The first design biases the study in favor of the hypothesis that there

exists a common morality. The second design biases the study against the

hypothesis.

These problems of potential bias and circularity can be controlled through proper

research design. It is reasonable to require, for example, that study subjects believe in

impartial application of a general principle of nonmaleficence. Acceptance of this


principle would serve as an inclusion criterion, and nonacceptance an exclusion

criterion. It is unimaginable that a morally committed person would not take this

principle seriously. This choice of a single general moral norm does not bias the

study by preselecting study subjects for the norms Gert and I judge to constitute the

common morality. Study subjects would not be screened for any norms other than

nonmaleficence and impartiality. Thepurpose of the study is todetermine the extent

to which cultural or individual differences emerge, or do not emerge, in study

subjects over a wide range of basic norms other than nonmaleficence and imparti-

ality, such as respect for autonomy, keeping promises, helping disabled persons, and

not punishing the innocent. The design of the study is not biased because these

norms are not entailed by, or an intrinsic part of, the principle of nonmaleficence or

a requirement of impartiality.

Should it turn out that the persons studied do not share one or more of the

norms that Gert or I hypothesize to comprise the common morality, this result

would show that study subjects have different conceptions about either the content

ofmorality or the concept ofmorality in the normative sense than those that Gert or

I have hypothesized. The study might even show that there is no commonmorality

or no concept of morality of the sort we claim. More optimistically, if the norms

that Gert and I hypothesize to constitute morality in the normative sense were

found to be shared by study subjects selected from a broad array of cultures, this

finding would constitute good initial evidence of the existence of the common

morality, though the study would undoubtedly need to be replicated.

No matter how careful in design, a single study is not likely to give us a

complete picture. Moreover, the results will likely not come packaged in the

distilled form a philosopher might prefer. It should be no surprise if an empirical

study of fundamental moral norms does not turn up exactly the same norms and

conclusions found in a conceptual analysis of fundamental moral norms in

philosophy. Conceptual analysis does not have to confront the complicated web

of moral beliefs found in social moralities, not to mention differences in indivi-

dual beliefs. Conceptual analysis and empirical investigation are such different

tools that they are likely to identify or discover somewhat different features in the

landscape of morals.

I have not claimed in this final section that empirical confirmation of the

hypothesis that there exists a common morality would constitute a normative

justification of the norms of the commonmorality, and I leave open whether, if the

common morality were found to exist, investigators would thereby be able to

establish its normative content. A well-designed empirical study has the potential

to justify or falsify the claim that some form of common morality exists, but it

would probably not discover every feature of the common morality. Still,

empirical study stands to inform us in the work of philosophical ethics, and we

ignore it at our peril.


CONCLUSION

In this essay I have tried to show that three types or strategies of justification of

claims about the common morality can be identified. My goal has been to classify

and provide a limited defense of the three methods. I have not offered a justifica-

tion of the common morality that satisfies the conditions of any of the three

forms, nor have I shown that Gert’s work is fatally flawed. Indeed, I have warmly

supported his view that the commonmorality merits themost careful attention by

philosophers, an attention it has rarely received, and never better than by Gert

himself.

References

Ainslee, Donald C. 2002. “Bioethics and the Problem of Pluralism.” Social Philosophy and

Policy 19, 1–28.

Arras, John. 2009. “The Hedgehog and Borg: Common Morality in Bioethics.” Theoretical

Medicine and Bioethics, as published online [unpaginated]. Available at: http://www.

springerlink.com/content/95x81214715395r2/.

Beauchamp, Tom L. 2003. “A Defense of the Common Morality.” Kennedy Institute of

Ethics Journal 13, 259–274.

Beauchamp, Tom L., and James F. Childress. 1st edition 1979, currently 6th edition 2009.

Principles of Biomedical Ethics. New York: Oxford University Press.

Bok, Sissela. 1995. Common Values. Columbia, MO: University of Missouri Press.

Brock, Dan W. 2001. “Gert on the Limits of Morality’s Requirement.” Philosophy and

Phenomenological Research 62, 435–440.

Clouser, K. Danner. 1995. “Common Morality as an Alternative to Principlism.” Kennedy

Institute of Ethics Journal 5, 219–236.

Daniels, Norman. 1979. “Wide Reflective Equilibrium and Theory Acceptance in Ethics.”

Journal of Philosophy 76, 256–282.

______. 1996a. “Wide Reflective Equilibrium in Practice.” In L. W. Sumner and J. Boyle,

eds., Philosophical Perspectives on Bioethics. Toronto: University of Toronto Press,

96–114.

______. 1996b. Justice and Justification: Reflective Equilibrium in Theory and Practice. New

York: Cambridge University Press.

______. 2003. “Reflective Equilibrium.” Stanford Encyclopedia of Philosophy. Online, first

published, April 28, 2003; accessed August 24, 2007 at http://plato.stanford.edu/entries/

reflective-equilibrium/.

DeGrazia, David. 2003. “Common Morality, Coherence, and the Principles of Biomedical

Ethics.” Kennedy Institute of Ethics Journal 13, 219–230.

Donagan, Alan. 1977. The Theory of Morality. Chicago: University of Chicago Press.

Engelhardt, H. Tristram, Jr. 1996. The Foundations of Bioethics. New York: Oxford

University Press.


http://www.springerlink.com/content/95x81214715395r2/

http://www.springerlink.com/content/95x81214715395r2/

http://plato.stanford.edu/entries/reflective-equilibrium/

http://plato.stanford.edu/entries/reflective-equilibrium/

Feigl, Herbert, 1950. “De Principiis Non Disputandum . . . ?” In Max Black, ed.,

Philosophical Analysis. Ithaca, NY: Cornell University Press, 119–156.

Frankena, William K., 1980. “What is Morality?” In his Thinking about Morality. Ann

Arbor: University of Michigan Press, Chapter 1.

Gert, Bernard. 1970. The Moral Rules: A New Rational Foundation for Morality. New York:

Harper and Row.

_____. 1988.Morality: A New Justification of the Moral Rules. New York: Oxford University

Press, 1988.

_____. 1998. Morality: Its Nature and Justification. New York: Oxford University Press.

_____. 2001. “Reply to Dan Brock.” Philosophy and Phenomenological Research 62, 466–470.

_____. 2004. Common Morality: Deciding What to Do. New York: Oxford University Press,

paperback edition 2007.

_____. 2005. Morality: Its Nature and Justification. New York: Oxford University Press.

_____. 2008. “The Definition of Morality.” The Stanford Encyclopedia of Philosophy. First

published Wed. April 17, 2002; here using the substantive revision of Mon. Feb. 11, 2008.

Pages in this article are determined by a file downloaded from the Stanford site, as

accessed March 15, 2009: http://plato.stanford.edu/entries/morality-definition/.

Gert, Bernard, and K. Danner Clouser. 1990. “A Critique of Principlism.” Journal of

Medicine and Philosophy 15, 219–36.

______. 1994. “Morality vs. Principlism.” In Principles of Health Care Ethics, ed. Raanan

Gillon and Ann Lloyd. London: John Wylie & Sons, 251–266.

______. 1999. “Concerning Principlism and Its Defenders: Reply to Beauchamp and

Veatch.” In Building Bioethics: Conversations with Clouser and Friends on Medical

Ethics, ed. Loretta Kopelman. Boston: Kluwer Academic, esp. 191–193.

Gert, Bernard, Charles M. Culver, and K. Danner Clouser. 1997. Bioethics: A Return to

Fundamentals. New York: Oxford University Press.

______. 1997. Bioethics: A Systematic Approach. New York: Oxford University Press, 2006.

______. 2000. “Common Morality versus Specified Principlism: Reply to Richardson.”

Journal of Medicine and Philosophy 25, 308–322.

Gert, Bernard, Ronald M. Green, and K. Danner Clouser. 1993. “The Method of Public

Morality versus theMethodofPrinciplism.” Journal ofMedicine and Philosophy 18, 179–197.

Hartland-Swann, John. 1960. An Analysis of Morals. London: George Allen & Unwin.

Herissone-Kelly, Peter. 2003. “The Principlist Approach to Bioethics, and its Stormy

Journey Overseas.” In Scratching the Surface of Bioethics, ed. Matti Hayry and Tuija

Takala. Amsterdam: Rodopi, 65–77.

Holm, Søren. 1994. “Not Just Autonomy—The Principles of American Biomedical Ethics.”

Journal of Medical Ethics 21, 332–338.

Lindsay, Ronald A. 2005. “Slaves, Embryos, and Nonhuman Animals: Moral Status and the

Limitations of Common Morality Theory.” Kennedy Institute of Ethics Journal 15,

323–346.

______. 2009. “Bioethics Policies and the Compass of Common Morality.” Theoretical

Medicine and Bioethics, as published online [unpaginated] at http://www.springerlink.

com/content/jh0l7t2034x72744/.


http://www.springerlink.com/content/jh0l7t2034x72744/

http://www.springerlink.com/content/jh0l7t2034x72744/

http://plato.stanford.edu/entries/morality-definition/

Locke, John. 2006. A Treatise Concerning the True Original Extent and End of Civil

Government, Second Treatise, Chap. 9. In The Longman Standard History of Modern

Philosophy, ed. Daniel Kolak and Garrett Thomson. New York: Pearson Longman, 276.

Mackie, John. 1977. Ethics: Inventing Right and Wrong. London: Penguin.

Rauprich, Oliver. 2008. “Common Morality: Comment on Beauchamp and Childress.”

Theoretical Medicine and Bioethics 29, 43–71.

Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press; revised

edition, 1999.

______. 1996. Political Liberalism. New York: Columbia University Press.

Richardson, Henry S. 1990. “Specifying Norms as a Way to Resolve Concrete Ethical

Problems,” Philosophy and Public Affairs 19, 279–310.

Ross, W. D. 1939. The Foundations of Ethics. Oxford: Oxford University Press.

Salmon, Wesley. 1966. The Foundations of Scientific Inference. Pittsburgh, PA: University of

Pittsburgh Press.

Strong, Carson. 2009. “Exploring Questions about Common Morality.” Theoretical

Medicine and Bioethics, as published online [unpaginated], section entitled

“Objections to Common Morality.” Available at http://www.springerlink.com/

content/tk66r0jw0u061804/.

Turner, Leigh. 1998. “An Anthropological Exploration of Contemporary Bioethics: The

Varieties of Common Sense.” Journal of Medical Ethics 24, 127–133.

______. 2003. “Zones of Consensus and Zones of Conflict: Questioning the ‘Common

Morality’ Presumption in Bioethics.” Kennedy Institute of Ethics Journal 13, 193–218.

Wallace, Gerald, and A. D. M. Walker. 1970. The Definition of Morality. London: Methuen

& Co., Ltd.

Wallace, K. A. 2009. “Common Morality and Moral Reform.” Theoretical Medicine and

Bioethics, published online [unpaginated], section on “Pragmatic Common Morality

and Moral Reform.” Available at: http://www.springerlink.com/content/

h6529372046v5220/.

Warnock, G. J. 1971. The Object of Morality. London: Methuen & Co.

Notes

1. K. Danner Clouser and Bernard Gert (1990); Ronald M. Green, Gert, and Clouser

(1993); and Clouser and Gert (1994).

2. Tom L. Beauchamp and James F. Childress (1979, as in the 6th edition 2009).

3. Good examples are John Arras (2009), Oliver Rauprich (2008), Peter Herissone-Kelly

(2003), and Søren Holm (1994). Other skeptical appraisals are given by H. Tristram

Engelhardt, Jr. (1996) and David DeGrazia (2003).

4. Norman Daniels (1979); Daniels (1996a); Daniels (1996b); and Daniels, Stanford

Encyclopedia (online, first published, April 28, 2003; accessed Aug. 24, 2007).

5. Gert rejects reflective equilibrium altogether—see Gert (2005), 380–381—whereas I

accept it for certain types of justification. See Beauchamp and Childress (2009), 381–387.

6. Gert (1970).


http://www.springerlink.com/content/tk66r0jw0u061804/

http://www.springerlink.com/content/tk66r0jw0u061804/

http://www.springerlink.com/content/h6529372046v5220/

http://www.springerlink.com/content/h6529372046v5220/

7. Gert (1988).

8. Gert (1998). Gert at last has a title not in need of revision, and the title, Morality: Its

Nature and Justification, is retained seven years afterward in the next edition (2005).

9. Gert (2004, paperback edition 2007). Gert (2005, vii) describes this commonmorality

book as “a short version of my theory.”

10. Gert and Clouser published, with Charles M. Culver (1997), a sustained criticism of

views that Childress and I had developed. This criticism is carried into the second edition of

the work (2006). See also Gert, Culver, and Clouser (2000); and Gert and Clouser (1999).

11. Clouser (1995). Clouser’s article was a later version of lecture delivered on March 6,

1994, at the Kennedy Institute of Ethics, Georgetown University. In this lecture he did not

use the language of “common morality”—thus suggesting that the mid-1994 to early-1995

period was pivotal for the change of language used by both Clouser and Gert.

12. Sissela Bok (1995).

13. Alan Donagan (1977).

14. My views on the common morality were originally formed by reflection on

Donagan’s book (1977). He had a distinct view of the commonmorality as heavily informed

by Kantian and Judeo-Christian thought—more of a western common morality theory

than a universal account. In retrospect I would have profited at the time more by a careful

assessment of Gert’s account of morality. My views wound up considerably closer to Gert’s,

and I came to regard his theory as far superior to Donagan’s.

15. Although there is only one universal commonmorality, there is more than one theory

of the common morality. For a variety of theories, see Donagan (1977); Gert (2004, paper-

back edition 2007); Beauchamp and Childress (2009); and W. D. Ross (1939).

16. Gert readily acknowledges disagreement among philosophers on the point and says that

there is not even “complete agreement concerningwhat counts as amoral rule.”Gert (2005), 13.

17. Gert (2005), 3; and see 159–161, 246–247.

18. I am indebted for my own views to Gert’s books and especially to his extensive article

in the Stanford Encyclopedia on the definition of morality (2008). The 2005 revision of

Gert’s masterwork covers the subject of “The Definition of Morality” in a single page

(13–14, but see also the preparatory work on p. 10).

19. Gert (2004, paperback 2007), 84.

20. Gert (2005), 29–33, 39–41, 181.

21. Gert (2005, 6) contains a more critical view of Hobbes, Kant, and Mill, who “provide

only a schema of the moral system” (not an “adequate account”).

22. Gert (2005), 18; (1998), 18.

23. Beauchamp (2003), 259–274. See also Rauprich (2008), 68; and K. A. Wallace (online

10 Feb. 2009), section on “Pragmatic Common Morality and Moral Reform.”

24. This particular formulation is indebted to Herbert Feigl (1950), 119–156; and Wesley

Salmon (1966).

25. Gert (2005), 13, 109, 246.

26. Gert (2005), 11–14.

27. See the sources referenced in Bok (1995), 13–23, 50–59 (citing several influential

writers on the subject).


28. G. J. Warnock (1971), esp. 15–26; also John Mackie (1977). I am indebted to both.

29. John Rawls (1971), esp. 1971, 20ff, 46–50, 579–580 (revised edition, 1999, 17ff, 40–45,

508–509). See also Rawls’s comments on reflective equilibrium in his later book (1996), esp.

8, 381, 384, and 399.

30. My view is that the norms are most plausibly understood as subject to what I call,

following Henry Richardson, specification—that is, the process of reducing the indetermi-

nate character of abstract norms and generating more specific action-guiding content.

Though constrained by standards of reflective equilibrium, specification is not identical to

reflective equilibrium. See Richardson (1990), 279–310.

31. Gert (2008), 2.

32. Gert (2008), 8 (italics added).

33. John Locke (2006), 276 (chap. 9.124).

34. Gert (2005), 7, 11–13, 219; Gert (1998), 160.

35. Gert (2008), 9.

36. Cf. Beauchamp and Childress (2009), 260–261, 361–363.

37. Gert (2005), 14.

38. Gert says that, “To say that moral rules apply to all moral agents means essentially the

same as to say that they apply to all rational persons” (2005, 112). In his Stanford

Encyclopedia article (2008, 11), he writes that “Since the normative sense of ‘morality’

refers to a universal guide to behavior that all rational persons would put forward for

governing the behavior of all moral agents, it is important to provide at least a brief account

of what is meant by ‘rational person.’ In this context, ‘rational person’ is synonymous with

‘moral agent’ and refers to those persons to whom morality applies. This includes all

normal adults with sufficient knowledge and intelligence to understand what kinds of

actions morality prohibits, requires, discourages, encourages, and allows, and with suffi-

cient volitional ability to use morality as a guide for their behavior.” See further Gert

(2005), 23, 136–147.

39. For example, see Gert (2004, paperback 2007), 7; and Gert, Culver, and Clouser

(1997), 7 (“the purpose of morality is to minimize the amount of evil or harm suffered by

those protected by morality”), and see also 62–68; Gert, Culver, and Clouser (2006), 89–93;

and Gert (2005), 17, 124–128.

40. Gert (2008), 13 (italics added here and in the paragraph following). In explicating the

normative sense of “morality,” Gert writes as follows (12): “That morality is a public system

does not mean that everyone always agrees on all of their moral judgments, but only that all

disagreements occur within a framework of agreement.”

41. For Gert’s tough line on confining moral obligation to the scope of a principle of

nonmaleficence, see Gert, Culver, and Clouser (1997), 7–8; and (2006), 11–13. For some

problems with Gert’s views and some concession on Gert’s part, see Dan W. Brock (2001),

435–440; and Gert, “Reply to Dan Brock” (2001), 466–470.

42. Such a philosophical analysis is advanced in John Hartland-Swann (1960) and

pursued in various articles in Gerald Wallace and A. D. M. Walker (1970), and also

William K. Frankena (1980).

43. Gert (2008), 13.


44. Gert (2005), 114–115. For more on the thesis that morality is unchanging, see Gert,

Culver, and Clouser (2006), 104.

45. Compare Ronald A. Lindsay (2005), 323–346.

46. I deal with this question in Beauchamp (2003).

47. Herissone-Kelly (2003), 65–77; and similarly Lindsay (2009), first section.

48. Gert, Culver, and Clouser (2003), 310.

49. Gert, Common Morality (2004, paperback edition 2007), v.

50. Gert, Common Morality (2004, paperback edition 2007), 8.

51. Gert (2005), 3.

52. Gert (2008), 4.

53. Gert (2005), 6.

54. Arras’s critique turns on Gert’s failure to deal adequately with questions of empirical

evidence and with whether it bears on his claims about morality. See “The Hedgehog and

the Borg” (2009), in the subsection “What is Common Morality?”

55. Gert (2005), 3.

56. Rauprich (2008), 47.

57. See Leigh Turner (2003); Turner (1998); Donald C. Ainslee (2002); Carson Strong

(2009), section entitled “Objections to Common Morality”; David DeGrazia (2003).

58. Beauchamp and Childress (2009), 392–394.


12

ONELIMINATING THEDISTINCTION BETWEEN

APPLIED ETHICS AND ETHICAL THEORY

“Applied ethics” has been a major growth area in North American philosophy in

recent years, yet robust confidence and enthusiasm over its promise are far from

universal in academic philosophy. It is considered nonphilosophical in

Germany, and has largely failed to penetrate British departments of philosophy.

Whether it has any intellectually or pedagogically redeeming value is still widely

debated in North America, where many who have tried to teach some area of

applied ethics for the first time have seen their courses grind to a halt in

irresolvable quandaries, practical exigencies, or unavailable technical informa-

tion. Consequently, some departments are at war over whether to hire or confer

tenure on those who teach these subjects. Many also remain unconvinced that

philosophical generalizations or theories can play any significant role in prac-

tical contexts. They even doubt both that applied ethics is truly philosophical

ethics and that philosophy should accommodate this blossoming field in its

curriculum.

Against the skeptical side of these opinions, I argue that no significant

differences distinguish ethical theory and applied ethics as philosophical activ-

ities ormethods. Philosophers do what they have always done since Socrates. I do

not argue that there are no differences in content, but this basis for a distinction

can be misleading because the essence of “applied ethics” is not captured by such

differences. Initially I argue for eliminating the distinction between applied

ethics and ethical theory. I then turn in subsequent sections to a particular

method—the case method. The methodology in this method is unsettled and

may be discipline specific, but it has played and will continue to play a sig-

nificant pedagogical role in “applied” ethics. I argue that this method has a

scholarly as well as a pedagogical value for philosophy and that the method can

be used to enhance traditional reflective uses of cases and examples in ethical

theory. Ethical theory, no less than applied ethics, can be enhanced by this

method.

THE STRANGE IDEA OF “APPLYING” ETHICAL THEORY

I shall not here attempt a deep conceptual analysis of “applied ethics,” but a basic

working understanding of the term is needed. Bernard Gert has provided a

standard definition: “the application of an ethical theory to some particular

moral problems or set of problems.” Gert maintains that applied ethics can be

concerned either with particular moral situations or with general moral problems

such as those found in political theory. For both the general and the particular,

ethical theory determines which facts count as morally relevant and which actions

can be considered morally justified or unjustified.1

For reasons that will become clear, I prefer not to use the term application, as

Gert does. While his definition is mainline, it is too restricted and does not

capture significant features of applied ethics. I will use “applied ethics” more

broadly to refer to the use of philosophical theory and methods of analysis to treat

fundamentally moral problems in the professions, technology, public policy, and

the like. Biomedical ethics, political ethics, journalistic ethics, jurisprudence, and

business ethics are fertile professional areas for such philosophical activity, but

“applied ethics” is not synonymous with “professional ethics” and does not

necessarily require an ethical theory. Staple problems such as the allocation of

scarce social resources, the entrapment of public officials, research on animals,

and the confidentiality of tax information are problems beyond those of profes-

sional conduct and relationships, and they may not be best addressed by a general

ethical theory.

APPLICATIONS OF CONTENT AND APPLICATIONS OF METHOD

Under this broad conception, there need not be a special philosophicalmethod for

applied ethics even when the work is fundamentally philosophical. Traditional

conceptions of the role or activity of philosophy typically undergo no major shift

to accommodate a new subject matter for philosophical reflection. New methods

of normative and conceptual analysis did not emerge with the development of the


philosophy of economics, for example, and nothing about our new interest in the

philosophy of sociobiology requires new methods for the philosophy of science.

Applied ethics, and applied philosophymore generally, are no different. “Applied

philosophers” do what philosophers have always done: They analyze concepts;

submit to critical scrutiny various strategies that are used to justify beliefs, policies,

and actions; examine the hidden presuppositions of various moral opinions and

theories; and offer criticism and constructive accounts of the moral phenomena in

question. They seek a reasoned defense of a moral viewpoint, and they use guides to

duty and virtue in order to distinguish justified moral claims from unjustified ones.

They try to stimulate the moral imagination of their students, promote analytical

skills, and prevent the substitution of purely personal attitudes or intuitions for a

reasoned and justified moral position. Prejudice, emotion, false data, false

authority, and the like are weeded from the premises of arguments, and moral

reasons are advanced and defended by recognizably philosophical modes of argu-

ment. Inquiry into the meta-ethics of applied ethics invites a wild goose chase if one

expects to find activities different from those philosophers have pursued since

Socrates submitted the great issues of his day to philosophical scrutiny.

When we turn frommethod to content, we expect and find differences. Courses

in business ethics or engineering ethics vary as much in content from courses in

ethical theory as courses in philosophy of sociobiology and philosophy of

psychology differ from those in the philosophy of science. In each case, however,

it is not an easy matter to say how or why they vary, and indeed they often do not

vary to the degree we might anticipate. Courses in philosophy of psychology

commonly begin with an introductory text in the philosophy of science, and

courses in business ethics typically begin with a major section on ethical theory.

Yet, as the courses proceed, differences in content from the “parent” field will

emerge: The concepts analyzed will be different, the topics treated will be dif-

ferent, the presuppositions examined will be different, and the principles

employed or investigated will vary.

Rather than analyze the word good, or other tasks of meta-ethics, philosophers

interested in applied ethics turn their efforts to the analysis of a range of concepts

including confidentiality, medicine, trade secrets, environmental responsibility,

euthanasia, authority, undue influence, free press, privacy, and entrapment. But

in noting these differences of content, we should note the similarities as well.

Some notions on this sample list play a significant role in both applied ethics and

ethical theory, with privacy and authority being obvious examples. In journals

such as Philosophy and Public Affairs, no sharp line of demarcation can be drawn

between ethical theory and applied ethics when such notions are examined. There

is not even a discernible continuum from theoretical to applied concepts or

principles. While there are discernible differences in content between ethical

theory and applied ethics, it is difficult to specify the respects in which they differ.

T H E D I S T I N C T I O N B E T W E E N A P P L I E D E T H I C S A N D E T H I C A L T H E O R Y 213

THE IDEA OF APPLYING GENERAL AND FUNDAMENTAL PRINCIPLES

A critic might object that there are more fundamental differences of content than

I have acknowledged. It seems widely presumed in contemporary ethics that the

“applied” part of “applied ethics” involves the application of basic principles

of ethical theory to particular moral problems, as Gert suggests. This outlook

promotes the thesis that ethical theory develops general and fundamental princi-

ples, virtues, rules, and the like—the “groundwork” in the German tradition—

while applied ethics treats particular contexts through less general, derived prin-

ciples, virtues, rules, judgments, and the like. There is no new or special ethics (of

general normative principles), but only the application of existing theories and

principles to specific cases and controversies. For example, principles of the

confidentiality of information in medicine, journalism, and political theory

derive from more general and fundamental principles of respect for persons,

promise keeping, and the like.

This thesis helps explain why some believe that applied ethics is doomed to

secondary importance. They think applied work fails to treat what is basic to

philosophical ethics—analysis of the most general and fundamental princi-

ples, virtues, presuppositions, and theories. The conventional wisdom behind

the distinctions in analytic philosophy between normative ethical theory and

applied ethics, in the Catholic tradition between general ethics and special

ethics, and in the German tradition between philosophical ethics and bureau-

cratic application seems in each case to be that a general normative ethics

collects the set of principles presupposed by or needed for application to

specific moral problems such as abortion, widespread hunger, and research

involving human subjects. These general principles are more fundamental

than other principles or rules because the latter can be derived from them,

while the general principles are not similarly dependent on these other

sources.

Presumably, the same general principles apply across many particular contexts.

If so, this is further witness to their status as fundamental. For example, one might

appeal to the same set of principles of justice to examine issues of taxation,

delivery of health care, criminal punishment, and reverse discrimination.

Similarly, general principles of veracity apply to debates about secrecy and

deception in international politics, misleading advertisements in business ethics,

accurate reporting in journalistic ethics, and the disclosure of the nature and

extent of an illness to a patient in medical ethics. Greater clarity about the

conditions in general under which truth must be told and when it may permis-

sibly be withheld should enhance understanding of what is required in all these

areas and would help us evaluate the moral actions of persons in a broad spectrum

of human activities.


Consider the example of an eight-chapter text in biomedical ethics that has the

following structure: The first chapter introduces morality and ethical theory, the

second chapter treats types of ethical theories, and the third through sixth

chapters treat the basic moral principles of respect for autonomy, nonmaleficence,

beneficence, and justice. The attempt throughout is to show how moral problems

in modern medicine and biology can be analyzed by appeal to these principles as

they pertain to specific problems. Many case studies and empirical studies are

examined, and the basic principles are used to argue for derivative rules such as

those governing valid third-party consent, permissible instances of letting die, and

medical paternalism.

Although the book mentioned is one I coauthored, I reject portions of this

interpretation of both the book and the method at work in applied ethics. This

interpretation neglects, for example, the role that case studies and empirical

studies play in the “derivation” of rules for biomedical ethics. The rules and

judgments proposed for acceptance are hammered out in this volume through an

adjustment of abstract moral principles to concrete cases and factual findings. For

example, rules about the disclosure of information are developed by appeal to the

principle of respect for autonomy, clinical cases of medical decision making, and

empirical studies of comprehension. While principles of respect for autonomy,

nonmaleficence, and beneficence all play significant roles in the justification of

rules and judgments about both disclosure and nondisclosure, psychological

information about and clinical experience in the practical problems of the

patient–physician relationship are no less important. Paradigm cases may sig-

nificantly shape rules, as may empirical studies of effective communication and

legal theories of liability that govern medical malpractice.

This description is typical of applied ethics contexts: Rules are shaped by (1)

moral principles, (2) empirical data, (3) paradigm cases, and (4) moral reflection

on how to put (1) through (3) into the most harmonious unit. I want to be clear,

however, about what I accept and what I reject in this interpretation. I accept the

view that certain broad fundamental moral principles give support to numerous

other principles, rules, and judgments (as part of their “foundation” or justifica-

tion); but support is not entailment, and the underlying ideas of generality,

application, and fundamentalness (as well as derivation) in this interpretation

can lead to a view that has more falsity and misleading statement than truth and

insight. Moral principles are rallying points for justification. They provide vital

premises, but before we render them the queen of the applied moral sciences, we

had better be sure they have the power to fulfill the assigned function.

Hegel was properly critical of Kant for developing an “empty formalism” in a

theory that preached duty for duty’s sake, but lacked the power to develop an

“immanent doctrine of duties” or any particular duties. All “content and speci-

fication” in a living code of ethics (and reflection on it) had been replaced by


abstractness.2 Although Kant is an extreme case in moral theory, Hegel’s point has

force for an evaluation of the abstract principles and rules in contemporary ethical

theory. They are relatively empty formalisms with little power to identify or assign

specific duties in almost any context in which moral problems are present.

Principles help us see the moral dimensions of such problems and help us get

started, but they are too weakened by their abstractness to give us particular duties

or to assign priorities among various assignable duties. For example, if nonma-

leficence is the principle that we ought not to inflict evil or harm, this principle

gives us but a weak start on themoral problem of whether active euthanasia can be

morally justified; and if we question whether physicians ought to be allowed to be

the agents of active euthanasia, we compound the problems further.

The whole of the corpus of writings of Kant, Mill, and Aristotle will not take us

very far in answering such questions, even if they do provide some important

insights that help us supply answers. I am not denying that principles of ethical

theory can be applied—in, for example, the way Gert suggests. Of course they can.

But they are applied as premises along with a great many other considerations;

moreover, the principles and their “derivative rules” are formulated, shaped, and

reshaped by the moral codes and judgments of everyday life. On this point Hegel

and the pragmatists have it over Kant.3

Much can also be learned in these discussions by the ingenuity that has

been devoted in recent epistemology to showing that foundationalist or level-

derivation theories misconstrue the structure and development of human knowl-

edge. Similar arguments could be deployed to show that morality is not structured

in a layered system of principles from which other items such as rules and

judgments are derived. If, as I claim, moral theories should not be modeled on

geometrical systems, but rather on the sciences whose principles have been, and

continue to be, shaped and modified by new data, unexpected cases and situa-

tions, predictive failures, and modified hypotheses, then we need to be very

careful with the notion of fundamental principles and its ally general ethical

theory, because they may do more to obscure the nature of moral thinking than

to illuminate it.

Consider an example from recent philosophical literature that shows how these

distinctions can prove to be more confusing than helpful. Norman Daniels has

devoted a major effort to addressing problems of macroallocation and health

policy, in an attempt to establish a theory of the just distribution of health care

resources.4 While he calls on Rawls at various points—the opportunity principle

being especially important—Daniels goes on to argue for principles specifying

resource allocation priorities and physician responsibilities that are in no respect

directly derived from Rawls or from anywhere else in ethical theory. Is Daniels’s

effort any more or less philosophical, fundamental, theoretical, or general than

Rawls’s? Despite what might be called his “applications” of Rawlsian arguments


and assumptions, is his work intrinsically different from Rawls’s own “applica-

tion” of Kant in order to develop his theory of “the basic structure”? Is Rawls an

applied philosopher because he adopts parts of Kant’s more general ethical

theory? Both Rawls and Daniels are writing about distributive justice, developing

philosophical theories, and using similar methodologies and content—all of

which I would prefer to call ethical theory in each instance and be done with

misleading descriptions in terms of “applied ethics.”

I do not, however, want to be associated with critics5 who have claimed that

Kantian and Rawlsian theories are irrelevant to and provide no helpful back-

ground for discussion of the moral problems under investigation in areas of

applied ethics. Daniels, Onora O’Neill, and others have shown by their labors

that this view of Rawls and Kant is mistaken.6 But Rawls himself has commented

that his is a strict compliance and ideal theory that does not provide major

applications to “pressing and urgent matters” such as regulation of the liberty

of conscience, restraining violence, just war, unjust regimes, and problems of

compensatory justice7:

[Innonideal theory]wemust ascertainhow the ideal conceptionof justice applies,

if indeed it applies at all, to caseswhere rather thanhaving tomake adjustments to

natural limitations, we are confronted with injustice. The discussion of these

problems belongs to the partial compliance part of nonideal theory. It includes,

among other things, the theory of punishment and compensatory justice, just war

and conscientious objection, civil disobedience andmilitant resistance. These are

among the central issues of political life, yet so far the conception of justice as

fairness does not directly apply to them.8

What we need at present in philosophy is a realistic estimate of the limited

power of theories such as Rawls’s to cultivate moral wisdom, while also appre-

ciating—as he does—how they can help nourishmore rigorous work in areas now

labeled “applied.”

We ought to turn our eyes in the other direction as well, as Alasdair MacIntyre

and Arthur Caplan have urged.9 Traditional ethical theory has as much to learn

from “applied contexts” as the other way around, as Caplan has noted:

There are many ways currently in vogue for constructing theories in

ethics. . . . [Rawls’s and Nozick’s] are not the only starting points for moral

theorizing. One might want to start with a rich base of moral phenomena

and, inductively, construct a moral theory that captures the richness and

complexity of moral life. . . . Far from being atheoretical, medical ethics and

its related enterprises in applied ethics compel attention to deep

philosophical questions about optimal research strategies in ethical


theorizing. . . . Theoreticians are always in danger of oversimplifying,

overidealizing, or underestimating the complexity of human behavior.10

Our greatest philosophical writings have been plagued by the latter problems,

and by the difficulty that no theory adequate to serve as the foundation for

application to concrete moral problems has ever been devised. Accordingly, the

conception of a unilateral direction to the flow of ethical knowledge, together with

a presumption of special professional expertise, is a posture of insularity and

arrogance that practitioners of ethical theory ought never to adopt.

CLOUSER’S ANALYSIS

Danner Clouser has offered arguments that expand on the approach found in

Gert’s definition and that offer a challenge to the conclusions I have defended.

Clouser accepts the distinction between applied ethics and ethical theory and

proposes that applied ethics must not be expected to have the “depth” that ethical

theory has. Bioethics and other applied areas are simply “ordinary morality

applied to new areas of concern.” New “derived rules” are yielded, but they are

not “special principles” because they are not new “basic moral rules.” Applied

work need not, he argues, have a “very sophisticated level of ethical theory or

foundations.” Instead, worthwhile applied ethics requires a detailed “ground

preparation” in “the field to which the ethics is being applied.” This ground

preparation includes a thorough acquaintance with the concepts, behavior,

beliefs, strategies, and goals of the field. New principles do not emerge; rather,

we “squeeze out all the relevant implications from the ones [ethics] already has.”

Clouser argues that philosophical “purists” who dismiss applied ethics as not “real

ethics” miss the rigor and careful analysis that goes into the ground preparation

and the squeezing procedure. They dismiss it, he suggests, because they “never

looked” closely either at the field or at what is accomplished in applied ethics.

A part of the problem is that the philosophical purist does not possess the

adequate ground preparation and so misses the essence of the application.11

Clouser is right about the purist, but his pedagogical model of applied ethics

causes him to miss its potential, and in some cases actual, nature and contribu-

tion. (“Perhaps it is the ‘ground preparation,’” he comments, “which is the

essence of applied ethics.”) Consider Clouser’s major example—“the issue of

informed consent.” Here, he says, we find that:

Endless articles say essentially the same thing; seldom is anything

conceptually new accomplished—that is, in terms of moral theory; but the


need for informed consent is uncovered and emphasized in new

circumstances and settings. At their intellectual and conceptual best,

biomedical ethics articles will interestingly explore or extend or challenge

ethical theories in light of the details of biomedicine, . . . so that the

implications of the old and familiar morality (whatever that might be) can

be more accurately “applied,” or more decisively rejected.12

The endless stream of boring articles cannot be denied, but a more accurate

characterization of the literature on informed consent is that “the old and

familiar morality” (“whatever that might be”) has not been skillfully applied

in philosophy to yield compelling insights into “the issue of informed consent.”

“Squeezing” is a misplaced metaphor; “drilling a dry hole” would be closer.

Disciplines other than philosophy have fared better in treating this moral

problem. In law, the health professions, and the social sciences13 a rich literature

with multiple methodological commitments and interdisciplinary affiliations

has been generated. To be sure, certain philosophical analyses of voluntariness,

coercion, and related concepts have well served those who have carefully ana-

lyzed the issue of informed consent. But the most helpful among these analyses

(Nozick’s work on coercion and Feinberg’s on voluntariness, say) have not in

even a circuitous fashion led to a theory of informed consent or to new norma-

tive rules of informed consent.

We are currently beginning to appreciate in moral theory that the major con-

tributions in this area have recently run from “applied” contexts to “general” theory

rather than from general to applied. In examining case law and other literature on

informed consent, philosophers have learned a great deal about moral problems in

the disclosure of information that are requiring us to rethink andmodify traditional

conceptions of truth telling, disclosure, lying, withholding information, under-

standing, consenting, contracts, and the like. To the extent that sophisticated

philosophical treatments of these notions are now emerging, they are due in

many cases to the examinations and refinements occurring in applied contexts,

and they are beginning to fortify rather than squeeze juice from“the old and familiar

morality.” The works produced by Sissela Bok, Gerald Dworkin, Bernard Gert and

Charles Culver, and the President’s Commission (which counted Dan Wikler, Dan

Brock, and Allen Buchanan among its staff philosophers)—all forged in a heavily

interdisciplinary environment—have pushed us along on issues surrounding

informed consent and truth telling more than any treatise in ethical theory in

decades. But we have been pushed even further along on informed consent by the

legal analyses presented by Alan Meisel, Jay Katz, Alex Capron, and the landmark

cases from Schloendorff to Canterbury. This contribution to the moral problems of

informed consent came not by the application of ethical theory, but rather by a

sensitive attention to the moral inadequacies of various legal approaches.


There was no model ethical theory in the pages of philosophers X, Y, and Z for

the advances made by the aforementioned philosophers and legal scholars. No

“sophisticated” background had been prepared fromwhich an application could be

squeezed into a moral theory of informed consent. This should not be surprising

once we understand the nature and limits of philosophy’s potential contribution.

The obligation to obtain informed consent in research and clinical contexts is

generally understood to be grounded in a principle of respect for autonomy, but

many issues about the nature and limits of this principle remain unsettled. For

example, what are the exact demands that the principle makes in the consent

context as to disclosure requirements? There is also much debate about whether

respect for autonomy demands informed consent where a relatively low level of risk

is attached. These questions are too specific for general normative ethics to hope to

determine answers from its repertoire of fundamental principles and virtues. More

important is that a philosophical account of respect for autonomy—not merely of

informed consent—is likely to be shaped by the practical realities and scholarly

studies in other disciplines that engage such problems as those of informed consent.

A model system of general normative ethics is so unavailable for this work that

no such application has been attempted. Even Gert’s TheMoral Rules could not be

milked to provide a theory or set of foundational rules in terms of which his

conceptual analyses and normative proposals for informed consent, competence,

and the like could be “derived” in his Philosophy in Medicine.14 As important as

The Moral Rules is to Gert’s subsequent work—and he is a model of a philosopher

attempting to make the proper connections—his “applied” work is no less “ethical

theory” than the treatment of moral rules. One reason applied ethics has flourished

of late, along with the revival of social and political philosophy, is because these

endeavors have shown more insight into the moral life than most of the previous

course taken by philosophical ethics in the twentieth century. I am not arguing that

“applied” work is displacing or should displace traditional ethical theory, but only

that recent work by the philosophers I have named should help displace some

shopworn ideas of what ethical theory is and can be. Until philosophy has learned

that “applied” work is not a separate category from ethical theory—a double-deck

conception of ethics, where one deck is the cheap bleachers—the profession will not

have understood the potential in this area of investigation.

It might be maintained that I am only stipulatively defining “ethical theory” to

suit my own conceptual preferences and that my argument does not exceed

stipulation, but this objection mistakes my meaning. My argument is that

worthy philosophical activity need not be modified or compromised by “applied

ethics.” I do not mean to alter themeaning of “ethical theory” or to stipulate a new

meaning. The new content supplied by these applied contexts will simply revive

an ethical theory troubled by an overdose of meta-ethics and the ever less exciting

wars between deontological and utilitarian theories.


TRADITIONAL CASE METHODS

I earlier invoked a distinction between method and content, and I want now to

return to the subject of method, in particular the case method. This method is

widely used in applied contexts, but is regarded by many philosophers as having

about the same degree of rigor and significance as teaching Kant’s categorical

imperative through audio-visual aids. I want to argue, however, that not only is

the case method a sound pedagogical technique with a distinguished history, but

also that it can be used to make important contributions to ethical theory. This

will lend further support to my arguments favoring elimination of the distinction

between ethical theory and applied ethics.

Only recently has applied ethics drawn attention in philosophy to the impor-

tance of case studies and the case method. There has been some attempt to say

why the method is important and how it can be effective. For example, Daniel

Callahan and Sissela Bok have suggested that “case studies are employed most

effectively when they can readily be used to draw out broader ethical principles

and moral rules . . . [so as] to draw the attention of students to the common

elements in a variety of cases, and to the implicit problems of ethical theory to

which they may point.”15 This method seems well suited to certain conceptions of

applied ethics—as a way to apply, draw out, and illustrate the general principles

explored in theory. If anything illustrates applications of theory, it might be

thought, this method does. It is not obvious, however, how to apply theory,

draw out principles, or find common elements. I argue in this section that

proper use of the case method undermines rather than supports the belief in

two worlds of theory and application.

This argument requires a brief examination of the history of reflection on

the use of case studies and the case method—a history to which philosophy

has yet to make a notable contribution. Yet this history has much to

teach us about discovering, modifying, and using ethical principles. I shall

begin with two methods of case analysis that have been widely employed

outside of philosophy departments in American universities for nearly a

century. This procedure will have the additional advantage of distancing

our investigations from philosophical preconceptions about theory and its

areas of application.

THE CASE METHOD IN LAW

The most extensive body of thought on strategies for analyzing cases is found in

the law, where “case law” establishes precedents of evidence and justification. The

earliest developments in the history of the legal use of the case method in law


schools occurred shortly after 1870, when Christopher Columbus Langdell revo-

lutionized academic standards and teaching techniques by introducing “the case

method” at theHarvard Law School.16 Langdell’s casebooks were composed of cases

selected and arranged to reveal to the student the pervasive meaning of legal terms,

as well as the “rules” and “principles” of law. He envisioned a dialectical or Socratic

manner of argument to reveal how concepts, rules, and principles are found in the

legal reasoning of the judges who wrote the opinions. The skillful teacher or legal

scholar presumably could extract these fundamental principles, much as a skillful

biographer might extract the principles of a person’s reasoning by studying his or

her considered judgments.

Langdell held that legal theory is a science, resting on an inductive method

modeled on the “scientific method” of the natural sciences. Teachers and students

were viewed as developing hypotheses in their dissection of cases and landing on

principles, just as any good scientist does. In the law, he argued, one extracts from

cases a select body of principles that frame the English common law. Even though

the many particulars in cases vary and their judicial conclusions are sometimes at

odds, the principles of judicial reasoning need not vary. Using this method, one

could presumably study exactly how far a principle validly extends, including

where it is invalidly applied.

Langdell’s dream of making the law a science and making its essence a few

abstract principles was eccentric by almost any standard. His vision tended to

suppress such matters as the politics of legislation, particular historical circum-

stances, jurisdictional variations, and actual legal practice. These were dismissed

as irrelevant, because historical development and political context were less

important in his conception than abstract principle. Langdell’s “principles” also

did not prove to be as uniform across court, context, or time as had been envi-

sioned, and incompatible and even rival theories or approaches of judges tended to

be controlling in many precedent cases. Scientific legal theory proved unattainable,

and the idea of abstract fundamental principles worked no better here than it has for

ethical theory. Eventually, legal realism shattered Langdell’s dream.

There are, however, important reasons why the case method, with appropriate

modifications, ultimately prevailed in American law schools and persists as a

basic model. Analysis of cases offers teachers a powerful and attractive tool for

generalizing from cases. Spanning the tangled web of details in particular cases are

something like what Langdell declared “the principles of law.” Judges must have

them in order to render opinions. Legal theory and its fundamental doctrines are

arguably found in and applied to cases. Moreover, such training in the case method

can sharpen skills of legal reasoning—both legal analysis and legal synthesis.

One can tear a case apart and then construct a better way of treating similar

cases. In the thrust-and-parry classroom setting, teacher and student alike have to

think through a case and its rights and wrongs. The method in this way prepares


the student for the practice of the law—the application to new cases—and not

merely for theoretical wisdom about it.

In the end, the method’s most enduring value is the way it teaches students to

distinguish principles and evidence. By examining cases, students learn which

courts are considered most adept at legal reasoning and where “the weight of

the evidence” lies. The case method has come to stand as a way of learning how to

assemble facts and judge the weight of evidence to enable the transfer of that

weight to new cases. This is done by generalizing andmastering the principles that

control the transfer, usually principles at work in the reasoning of judges.

THE CASE METHOD IN BUSINESS

The casemethod has also dominatedmany business schools, where it has never been

hailed as a science or even as a single uniform method. Rather, the emphasis has

been on deliberation and decision making—making analyses, weighing competing

considerations, and reaching a decision in complex and difficult circumstances.17

Judgment is taught rather than doctrine, principle, or fact. Cases that involve

puzzles and dilemmas without any definitive solution along lines of principles or

rules are therefore preferred over those that fail to present a difficult dilemma. Cases

are not primarily used to illustrate principles or rules, because the latter are

inadequate to achieve the goals of the enterprise. Their inadequacy is part of the

problem from the perspective of business: There are no reliable rules that can be

generalized and transferred. There is no parallel to “the law.” The objective, then, is

to develop a capacity to grasp problems and to find novel solutions that are

contextually workable. To adapt Gilbert Ryle’s distinction: Knowing how is more

prized than knowing that.18

Business and law are not the only fields that employ a case-studies methodology

from which philosophers can learn. Case studies in the history of science have been

used by historians, social scientists, and philosophers in order to rationally recon-

struct conceptions of scientific methodology and theory. The reconstruction itself

can be a form of normative theory.19 Even in a field seemingly as remote from

“theory” as textual bibliography, series of case studies have been used both to

develop principles for establishing authoritative critical texts for classical writings

and to caution against overly rigid principles. A leading bibliographer—Philip

Gaskell—has argued that rather than sticking with a set of a priori bibliographical

principles, it is best to adapt contingently held principles for each particular case to

which they are to be applied.20 The history of recent bibliography has shown that

attachment to a preconceived theory or set of principles in textual bibliography has

produced utterly bizarre outcomes when applied to some texts.


There are, of course, dangers in transferring the case methods in law, business,

and other fields to philosophical ethics if, in their examination, the cases are

uninformed by ethical theory. Not much is drearier than a tedious and unre-

warding exposure to the moral opinions of those who are ignorant of theoretical

materials that transmit sustained philosophical reflection. However, this fact does

not lead to the conclusion that studying cases must be informed by theory while

theory can remain in isolation from modification by case study.

THE CASE METHOD IN PHILOSOPHY

The case method has made minor inroads in philosophy courses in recent years

because cases assist in focusing and dramatizing moral problems while locating

them in real-life rather than hypothetical circumstances. This developmentmay at

first glance seem novel, and perhaps even pedagogically revolutionary, but

Socrates knew it well. His own case of struggling with obedience to the state,

and whether or not to escape from its clutches, is a classic source of philosophical

reflection emerging from a particular occasion. But this is no isolated instance of

the Socratic style of reflection. While he published no ethical theory as we now

frame it, everyone recognizes that he taught ethical theory by using a method that

richly deserves the association with his name that it currently enjoys.

It takes little acquaintance with that method to know that Socrates served as a

midwife by eliciting from the student reflection, insight, and both theoretical and

practical judgment. The method starts with a profession of ignorance (not a

parading of theory) and proceeds to pointed questions that eventuate in proposed

principles or definitions. The latter are tested by hypotheses, then modified and

tested further until insight into principles or definitions is achieved.

Modifications specifically involve repeated appeals to cases, as in the Meno

when Socrates is discussing whether virtue can be taught. In his discussion with

Anytos and Meno, a string of cases of Sophists and fathers and sons is detailed to

get at this question, and at the end of his analysis of the various cases, Socrates says

that he fears that virtue cannot be taught. His cases are short, but they are

constantly at work in shaping his general moral views.

Philosophy has generally neglected to adapt something like the style of case

method successfully used in law and business to the philosophical method

practiced no less successfully by Socrates. Philosophers have generally preferred

in their writings to confine their attention to “hypothetical cases” or “paradigm

cases” of a few sentences. Usually they are explicitly fashioned to support a theory,

although the method of counterexample improves this defect by presenting “the

hard cases.” The best practitioners of this method do adjust the theory through the


counterexamples. While I think philosophy could have been stimulated by using

the more complex cases presented in business schools, I have no quarrel with or

reservation about an example–counterexample method of philosophizing. I do,

however, quarrel with those who associate the use of more extensive case analyses

with an “applied” rather than “theoretical” orientation to ethics. They are under the

false assumption that theory is not extracted from the examination of cases but only

applied to cases. My view, by contrast, is that Socrates was on the right track in using

cases to develop and modify theory, as was Hegel in his reflection on the cultural

roots of moral theory. Cases provide not only data for theory but also its testing

ground. Cases lead us to modify and refine embryonic theoretical claims, especially

by pointing to inadequacies in or limitations of theories.

A more extensive case analysis in philosophy might be joined with other philo-

sophical methods such as John Rawls’s account of reflective equilibrium, which I

shall modestly redescribe for present purposes. In developing a normative ethical

theory, he argues, it is appropriate to start with the broadest possible set of our

considered judgments about a subject such as justice, and to erect a provisional set

of principles that reflects them. These principles can then be pruned and adjusted

by bringing cases under them. If, for example, some problems in “applied ethics”—

say, deceptive advertising or environmental responsibility—are under examina-

tion, widely accepted principles of right action (moral beliefs) might be taken, as

Rawls puts it, “provisionally as fixed points,” but also as “liable to revision.”21

Paradigm cases of what clearly are right courses of action could first be examined,

and a search could then be undertaken for principles consistent with our judgments

about these paradigm cases. These principles could be tested by reference to other

paradigm cases and other considered judgments found in similar cases in order to

see if they yield counterintuitive or conflicting results. Through this process, moral

theories and principles could be made to cohere with considered judgments about

particular cases. That is, general ethical principles and particular judgments can be

brought into equilibrium. The principles and theory are justified by the process.

Presumably, the more complex and difficult the cases that force revisions, the richer

the resultant theory will be.

Although Rawls uses this strategy as a supplement to his Kantian moral theory,

reflective equilibrium has been subjected to criticism in recent ethics on grounds

that it merely brings judgments and intuitions into coherence, without thereby

providing theoretical justifications. In effect, this amounts to the standard objec-

tion to coherence theories of truth. However, the account can probably be

formulated to overcome these difficulties,22 and in any event I am not arguing

that all of moral theory should be derived from this form of analysis. I am arguing

only that moral thinking resembles other forms of theorizing in that hypotheses

must be tested, buried, or modified through relevant experiments. Principles can

be justified, modified, or refuted—and new insights gained—by examination of


cases that function as experimental data. Likewise, our developed principles allow

us to interpret the cases and arrive at moral judgments in a reflective and

accomplished manner. This is the dialectic in moral thinking that Socrates

symbolizes. One promise of the case method is that philosophers could increase

the quality and applicability of their ethical theories by more careful attention to

complicated, real-world cases that provide opportunities for testing the scope,

consistency, and adequacy of those theories.

There is no single method delineating how cases are to be used in either law or

business, and we should not expect an entirely uniform conception in philosophy.

I have presented a programmatic proposal about the promise of the case method

for pedagogical as well as theoretical objectives. It is an undervalued tool for

reflection that stands to enhance ethical theory more than philosophers presently

seem to appreciate. If philosophy can learn to move easily from case studies and

issues in private, professional, and public ethics to various more traditional

concerns in ethical theory—while training students to reason both without fear

of theoretical “irrelevancies” and without disdain for practical applications—the

profession will have taken to heart the lesson that Socrates long ago taught.

However, as warmly as I endorse the case method, we must not expect too

much of it, just as we must not expect too much of moral philosophy generally

in the way of final methods or solutions to moral problems. Philosophical ethics

provides reasoned and systematic approaches to moral problems, not finality.

Notes

1. Bernard Gert, “Licensing Professions,” Business and Professional Ethics Journal 1

(Summer 1982): 51–52.

2. G. W. F. Hegel, Philosophy of Right, trans. T. M. Knox (Oxford: Clarendon Press,

1942), 89–90, 106–107.

3. On the promise of pragmatism for applied ethics, see Andrew Altman,

“Pragmatism and Applied Ethics,” American Philosophical Quarterly 20 (April 1983):

227–235.

4. See, for example, Daniels’s “Health Care Needs and Distributive Justice,” in

Philosophy and Public Affairs 10 (Spring 1981). Daniels’s book on the subject is forthcoming

from Cambridge University Press (1985).

5. Cf. Ched Noble, “Ethics and Experts,” The Hastings Center Report 12 (June 1982): 7–10,

with responses by four critics.

6. See Onora O’Neill, “The Moral Perplexities of Famine Relief,” in Tom Regan,

ed., Matters of Life and Death (New York: Random House, 1980), pp. 260–298, esp.

section IV.

7. John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971), 8–9,

245–247.

8. Ibid., 351.


9. Alasdair MacIntyre, “What Has Ethics to Learn from Medical Ethics?” Philosophic

Exchange 2 (Summer 1978): 37–47; Arthur L. Caplan, “Ethical Engineers Need Not Apply:

The State of Applied Ethics Today,” Science, Technology, and Human Values 6 (Fall 1980):

24–32, and reprinted in S. Gorovitz, et al., eds., Moral Problems in Medicine, 2nd ed.

(Englewood Cliffs, NJ: Prentice-Hall, 1983).

10. Caplan, “Ethical Engineers Need Not Apply,” in Gorovitz, et al., 40–41.

11. K. Danner Clouser, Teaching Bioethics: Strategies, Problems, and Resources (Hastings-

on-Hudson, NY: The Hastings Center, 1980), 53–57; “Bioethics: Some Reflections and

Exhortations,” The Monist 60 (1977): 47–61; and “Bioethics,” Encyclopedia of Bioethics,

Vol. I, ed. Warren T. Reich (New York: Free Press, 1978).

12. Clouser, Teaching Bioethics, 55.

13. For the best philosophy has had to offer, see Gerald Dworkin, “Autonomy and

Informed Consent,” in President’s Commission for the Study of Ethical Problems in

Medicine and Biomedical and Behavioral Research, Making Health Care Decisions, Vol. 3

(Washington, DC: U.S. Government Printing Office, 1982), 63–82.

14. Philosophy in Medicine: Conceptual and Ethical Issues in Medicine and Psychiatry

(New York: Oxford University Press, 1982) contains surprisingly little of the moral theory

developed in The Moral Rules (New York: Harper Torchbooks, 1975). The application

comes almost exclusively in discussions of paternalism. See Philosophy inMedicine, 130–163.

(Chapter 2 in this book also relies on an account of rationality developed in the earlier

book.) See also pp. 42–63 for discussions of informed consent and competence that rely

surprisingly little on the earlier book. The best example of his attempts at application is

Gert’s “Licensing Professions,” which is developed almost entirely within the framework of

TheMoral Rules. This essay suffers, in my judgment, from an overdose of ethical theory and

a miniscule treatment of licensing professions (as Gert hints, on pp. 57 and 59). The essay

serves, however, to show how difficult it is to bring a systematic moral theory to bear on

such concrete problems without a total immersion in the surrounding “nonphilosophical”

material.

15. Daniel Callahan and Sissela Bok, The Teaching of Ethics in Higher Education

(Hastings-on-Hudson, NY: The Hastings Center, 1980), 69. See also Thomas Donaldson,

“The Case Method,” in his Case Studies in Business Ethics (Englewood Cliffs, NJ: Prentice-

Hall, 1984), 1–12.

16. Langdell’s first casebook on Contracts is treated in Lawrence M. Friedman, A History

of American Law (New York: Simon and Schuster, 1973), 531f. The general account of the

case method in this section is indebted to this source, and also to G. Edward White, Tort

Law in America: An Intellectual History (New York: Oxford University Press, 1980).

17. See M. P. McNair, ed., The Case Method at the Harvard Business School (New York:

McGraw-Hill, 1954).

18. Gilbert Ryle, “Knowing How and Knowing That,” Proceedings of the Aristotelian

Society XLVI (1945–1946): 1–16.

19. See Richard M. Burian, “More Than a Marriage of Convenience: On the

Inextricability of History and Philosophy of Science,” Philosophy of Science 44

(1977): 1–42.


20. Philip Gaskell, From Writer to Reader (Oxford: Oxford University Press, 1978), and

A New Introduction to Bibliography (Oxford: Oxford University Press, 1972).

21. John Rawls, A Theory of Justice, 20f.

22. See Norman Daniels, “Wide Reflective Equilibrium and Theory Acceptance in

Ethics,” Journal of Philosophy 76 (1979): 256–282. See also Altman, “Pragmatism and

Applied Ethics,” 229–231.


13

DOES ETHICAL THEORY HAVE A FUTURE

IN BIOETHICS?

The last 25 years of published literature and curriculum development in bioethics

suggest that the field enjoys a successful and stable marriage to philosophical

ethical theory. However, the next 25 years could be very different. I believe the

marriage is troubled. Divorce is conceivable and perhaps likely. The most philo-

sophical parts of bioethics may retreat to philosophy departments while bioethics

more generally continues on its current course toward a more interdisciplinary

and practical field.

I make no presumption that bioethics is integrally linked to philosophical

ethical theory. Indeed, I assume that the connection is contingent and fragile.

Many individuals in law, theological ethics, political theory, the social and

behavioral sciences, and the health professions carefully address mainstream

issues of bioethics without finding ethical theory essential or breathtakingly

attractive. This is not surprising. Moral philosophers have traditionally for-

mulated theories of the right, the good, and the virtuous in the most general

terms. A practical price is paid for this theoretical generality1: It is often

unclear whether and, if so, how theory is to be brought to bear on dilem-

matic problems, public policy, moral controversies, and moral conflict—

which I will here refer to as problems of practice. By “problems of practice”

I mean the moral difficulties and issues presented in health policy and the

health professions when decisions must be made about a proper action or

policy.

Historically, no moral philosopher has developed a detailed program or

method of practical ethics supported by a general ethical theory. There is no

well-developed conception of practical ethics or its relation to theoretical ethics.

Many philosophers—including Peter Singer and Richard Hare, as utilitarians, or

Alan Donagan and Alan Gewirth, as nonutilitarians—could be presented as

counterexamples, but I am skeptical of this assessment. None has offered a

detailed statement of a method for moving from a general theory to the practical

implications of the theory; nor has any developed a fine-grained conception of

notions such as “practical ethics” and “practical method.” Further, although

many moral philosophers are, at present, actively involved in problems of bio-

medical ethics such as clinical and corporate consulting, policy formulation, and

committee review, questions are open regarding what their role as moral philo-

sophers should be and whether they can successfully bring ethical theories and

methods to bear on problems of practice.

I will use the term ethical theory to refer to normative philosophical theories of

the moral life and philosophical accounts of the methods of ethics. Writers in

bioethics such as Peter Singer, Bernard Gert, and Norman Daniels have offered or

discussed ethical theories, method in ethics, and practical ethics in the relevant

sense. While my definition of “ethical theory” is narrow, it should be suitable for

present purposes. I do not assume that only philosophers develop normative

theories or that the theories of philosophers occupy a privileged position. I simply

restrict the discussion to philosophical theories. I also will not discuss the value of

philosophical argument as a resource for the treatment of specific issues in

bioethics. For example, I will not discuss whether philosophical ethics can be

brought to bear on topical questions such as whether theories of distributive

justice and methods of philosophical analysis are currently contributing to the

discussion of population ethics and global inequities in health care, which is a

promising but still developing area of philosophical investigation.

My concerns are principally with the types of theory andmethod that have been

under discussion in bioethics in the last quarter-century. Three interconnected

areas have been prominent: (1) general normative moral theories (from utilitarian

and Kantian theories to principlism, casuistry, virtue ethics, feminist ethics,

particularism, etc.); (2) moral and conceptual analyses of basic moral notions

(informed consent, the killing/letting-die distinction, etc.); and (3) methodology

(or how bioethics proceeds—e.g., by use of cases, narratives, specified principles,

theory application, reflective equilibrium, legal methods, etc.). I leave it an open

question whether (2) or (3) can be successfully addressed without addressing (1),

an unresolved problem in ethics.

I will question philosophy’s success in all three areas, laying emphasis on its

failure to connect theory to practice. In order to survive and flourish in the

interdisciplinary environment of bioethics, work will have to improve in


philosophical quality and will have to directly engage problems of practice such as

whether public laws need modification, whether consensus is an acceptable

standard in committee review, ways in which the system for protecting human

subjects needs reform, and how decision-making rights are best protected in the

health care system. I believe that bioethics needs philosophical theory and stands

to profit from it, but better conceptions of method and applied argument are also

needed.

In assessing the contemporary literature and how it should change, I will

confine attention to three areas of the intersection between bioethics and ethical

theory: cultural relativity and moral universality, moral justification, and con-

ceptual analysis. In each case I will argue that philosophers need to develop

theories and methods that are more closely attuned to practice. The future of

philosophical ethics in interdisciplinary bioethics may turn on whether this

challenge is met.

CULTURAL RELATIVITY AND MORAL UNIVERSALITY

Prior to the early 1970s, there was little reason to conceive medical ethics or research

ethics as intimately connected to philosophical ethical theory. However, the 1970s

brought fears in medicine that the cement of its moral universe—the Hippocratic

tradition—was woefully inadequate for clinical ethics and inappropriate for

research ethics. A felt need arose for moral norms that could guide investigation

of prevailing clinical and research practices. Frameworks drawn from moral philo-

sophy suggested that bioethics could be given universal and principled moral

foundations as well as practical methods of inquiry. It was reassuring that med-

ical-moral judgments are not like leaves blowing in the shifting winds of cultural

relativism, subjective feelings, institutional commitments, and political voting.

Thus, there arose in the late 1970s and early 1980s the idea that universal

principles drawn frommoral philosophy could be made theoretically interesting

and practically relevant in medical ethics.2

Internal Challenges to Ethical Theory

For some years thereafter this assumption went largely unchallenged in the

literature of bioethics. However, approaches that invoke universal principles

began to be vigorously challenged in the late 1980s. Two prominent moral

philosophers, Danner Clouser and Bernard Gert, wrote an article that captured

widely shared concerns about appeals to universal principles that had emerged in

bioethics. At approximately the same time, Albert Jonsen and Stephen Toulmin

D O E S E T H I C A L T H E O R Y H A V E A F U T U R E I N B I O E T H I C S ? 231

introduced objections to all principles that were interpreted as rigid universal

principles. They judged such principles unsuitable for bioethics.3 Soon there

appeared case-based approaches, rights-based approaches, virtue-based

approaches, feminist approaches, and assorted attempts to show that one

approach to moral thinking in bioethics is superior to others. Some approaches,

though not all, seemed to incorporate an ethical relativity or subjectivity that

mitigated the hope for universal norms.

After two decades of clashes of “theory,” this discussion has helped clarify many

matters, but there is no need to embrace only one of the frameworks that have

seemed in intractable conflict. One can without inconsistency embrace principles,

rules, virtues, rights, narratives, case analysis, and reflective equilibrium. Bioethics

is no longer profiting from these discussions of theory conflict. Moreover, these

disputes have long had little to do with clinical and research practices.

I expect this “theory” part of the landscape of bioethics to vanish soon, because it is

serving nouseful purpose.What is enduring, as a legacy, is how to handle the original

hope for universal principles, rights, or virtues—and how to make these norms

relevant to practice. Those, including me, who defend universal norms are currently

under attack on grounds either that we make inadequately defended claims of

universality or that we claim too much universality. This issue has implications for

practice in both clinical medicine and research. If there are no transcultural, transin-

stitutional norms, then it appears that any set of rulesmay be as viable as any other set

of rules and that external criticism of rules has no substantive basis.

External Challenges to Ethical Theory

Before coming to philosophical treatments of this problem, I will discuss two

critics of universal norms who hail from disciplines other than philosophy. They

propose that universalist ethics be replaced in bioethics by pluralism, multi-

culturalism, postmodernism, or a practice-based account of morals. The first

critic is Nicholas A. Christakis:

The proper approach to ethical conflict recognizes that culture shapes (1) the

content of ethical precepts, (2) the form of ethical precepts, and (3) the way

ethical conflict is handled. Medical ethics may be viewed in cross-cultural

perspective as a form of “local knowledge.” . . . The hallmarks of such an

approach are ethical pluralism and humility rather than either ethical

relativism or universalism.4

Christakis commends the approach of Barry Hoffmaster and supports a

“contextualist view of medical ethics.” He endorses the surprisingly influential


thesis of Renee Fox and Judith Swazey (a thesis advanced without evidence or

argument) that mainstream ethical theory—based in “AngloAmerican analytic

philosophical thought”—has led “American bioethics” into “an impoverished

and skewed expression of our society’s cultural tradition.”5

The second critic is Leigh Turner, who agrees that contextualism is a worthy

model. He argues that the

presuppositions that many bioethicists seem to find persuasive fail adequately

to recognize the moral conflicts that can be found in a multiethnic,

multicultural, multifaith country such as the United States. As a conceptual

package, [these presuppositions] lead to a style of moral reasoning that

drastically overestimates the ability of bioethicists to “resolve” moral issues

because they exaggerate the extent to which morality exists in a stable, orderly

pattern of wide reflective equilibrium. This criticism is relevant to the work of

principlists [Beauchamp and Childress], casuists [Jonsen and Toulmin], and

the common morality approach of Gert, Culver, and Clouser. . . . My chief

criticismofuniversalist and generally ahistorical accounts [is that they] . . .have

not made a persuasive argument in support of the claim that there are cross-

cultural moral norms supporting a “universal” common morality.6

These two statements are notable for their dismissiveness, even hostility, with

regard to mainstream ethical theory. Only contextualism, which is more antitheory

than theory, survives unbesmirched. Cultural anthropology ("an ethnography of

the practice ofmorality inmedical contexts"7) is judged to be better positioned than

ethical theory to address questions of cultural conflict, transcultural bioethics, and

shared principle. These representations seem to rest on both misunderstandings of

ethical theory and an extraordinary faith in a vague and undefended contextualism.

However, my objective is not to assess the conclusions that Turner and Christakis

reach about universalist ethical theory. My intention is to focus on their reserva-

tions about ethical theory to see what can be learned from them.

There is much to be learned. The problem is that the philosophical literature in

bioethics on universal principles and rationally required behavior is often bereft of

a compelling defense and of a clear statement of implications for practice. I will

now argue that the literature fails to engage challenges presented by other dis-

ciplines or to connect theory to practice and practice to theory.

Moral Philosophers on the Threat of Relativism

Philosophers in bioethics often condemn a specific cultural practice more by

assuming the wrongness of the distasteful practice than by rebutting it through a

careful analysis or a general theory. But if their claims to differentiate warranted


and unwarranted practices are not grounded in a defensible moral account, then

relativism, pluralism, and multiculturalism become more, rather than less,

attractive.

As an example, I turn to the best work thus far published in bioethics as an

attempt to refute relativism and support universalism: Ruth Macklin’s Against

Relativism: Cultural Diversity and the Search for Ethical Universals in Medicine.8

This book is motivated by her understandable dismay over literature in

bioethics in which it is asserted that all norms are socially constructed and

that context is everything. She is flabbergasted that this view has become

fashionable. In response, Macklin endorses “fundamental principles” that

“can be universally applied” and proposes valid cross-cultural judgments. She

grants that many norms are relative to the moralities of particular cultures while

insisting that violations of basic human rights are universally proscribed.

Macklin deftly handles the thesis that acceptance of universality does not

undercut acceptance of appropriate measures of particularity. She argues,

rightly in my view, that we can consistently deny universality to some justified

moral norms while claiming universality for others. Cultural diversity therefore

does not entail moral relativism.

Macklin’s book is the most thorough treatment of relativism in bioethics by a

universalist. It contains compelling cases based on her experience in diverse

cultures. She presents stimulating reflections on moral progress, cultural

imperialism, basic human rights, cultural sensitivity, and the like. She goes

out of her way to make her views accessible to an interdisciplinary audience

and to avoid writing, as she says, in an “ahistorical” or “acontextual” manner.

Her book displays experience with and practical wisdom about many problems

in bioethics.

But is this book an example of bioethics grounded in ethical theory, Macklin’s

native discipline? Does she either bring a well-defended ethical theory to bear on

practice or use practice as the springboard to ethical theory? And does she

defend her universalism by philosophical argument? In a notably negative

assessment of Macklin’s book, Daniel Sulmasy rightly argues that Macklin

does not attempt to defend the principles at the core of her arguments through

ethical theory, does not distinguish her principles from certain exceptionless

moral rules that she rejects, and sometimes relies more on moral intuitions than

on universal principles:

She never makes any arguments for where [her] principles come from or why

they should be considered universal, but for her, these are clearly cross-

cultural truths. Nonetheless, distinguishing between fundamental moral

principles and more specific moral rules, she claims that such universal

truth only exists at the abstract level of principles. She states unequivocally,


“I don’t believe that exceptionless moral rules exist.” Thus, there is plenty of

room for moral rules to vary among cultures and individuals. . . . I looked

forward to understanding how a more liberal author might ground an anti-

relativist position. However, Macklin has provided no such ground. The

book is almost devoid of philosophical argument. . . . Those looking for

ethical justification for their positions should look elsewhere.9

Sulmasy provides criticisms of Macklin that—independent of the specific

content and issues in her book—capture my sense of deficiencies in the use and

nonuse of ethical theory in bioethics. The defects he sees in her book are problems

that I find ubiquitous in bioethics: There is a failure to engage the arguments of

the opposition (more dismissal than confrontation by argument)10 and an

absence of theory to support central conclusions.

At the same time, Macklin is understandably distressed by multiculturalism’s

current attractiveness. She finds that pluralism, relativistic multiculturalism,

and contextualism are rarely supported by careful argument or grounded in an

ethical theory. Everything that Sulmasy says in critical evaluation of Macklin’s

book could be directed at some of the leading defenses of contextualism. This

debate over universalism, pluralism, and relativism—like many discussions in

bioethics—needs a more careful grounding in a literature controlled by argu-

ment that seriously engages opposition cases and disagreement.11

No less important, I shall now argue, is the lack of a serviceable literature on

moral justification. I have proposed that philosophers who proceed from a

thorough understanding of practice often fail to produce or to ground their

conclusions in an ethical theory. I now reverse the direction of argument and

maintain that philosophical theories commonly fail to reach down to problems of

practice.

MORAL JUSTIFICATION

Philosophers are, by reputation and in self-conception, experts on the theory of

justification. No issue is more central to the work of a philosopher. However,

one would not know this by looking at the literature of ethical theory as it has

been used in bioethics. It is often doubtful whether and, if so, how philosophers

in bioethics conceive of justification and how they attempt to justify their

claims. It can be unclear whether the justification is normative or nonnorma-

tive, what the method of justification is, and even whether a justification is being

offered. Detailed discussions of the nature and standards of justification are

rare, and, when they do appear, are generally unhelpful to other philosophers

and confusing to an interdisciplinary audience.


The Reflective Equilibrium Model

The best work by a figure in bioethics on the theory of justification, in my

judgment, is that of Norman Daniels.12 Using Rawls’s celebrated account of

“reflective equilibrium,”13 Daniels argues that moral principles, theoretical

postulates, and moral judgments about particular situations should be rendered

coherent. His method requires consolidation of an array of moral-theoretical

beliefs as well as principles and concepts drawn from diverse ethical theories,

ethical traditions, and cultures. Here is Daniels’s streamlined statement of the

method:

All of us are familiar with the process of working back and forth between our

moral judgments about particular situations and our effort to provide

general reasons and principles that link those judgments to ones that are

relevantly similar. Sometimes we use this process, which is what Rawls calls

“narrow reflective equilibrium,” to justify our judgments, sometimes our

principles. We can still ask about the principles that capture our considered

judgments: Why should we accept them? To answer this question, we must

widen the circle of justificatory beliefs. We must show why it is reasonable to

hold these principles and beliefs, not just that we happen to do so. Seeking

wide reflective equilibrium is thus the process of bringing to bear the

broadest evidence and critical scrutiny we can, drawing on all the different

moral and nonmoral beliefs and theories that arguably are relevant to our

selection of principles or adherence to our moral judgments. Wide reflective

equilibrium is thus a theoretical account of justification in ethics and a

process that is relevant to helping us solve moral problems at various levels

of theory and practice.14

This is sweet reason, but is it more than common knowledge ("all of us are

familiar")? Is reflective equilibrium a theory that, as Daniels’s last sentence

implies, can be brought to bear on practice? If so, whose practice and in which

ways? Even basic notions such as “evidence” and “selection of principles” are

underanalyzed. The more Daniels has elaborated and deepened the theory, the

less accessible it has become to those not heavily invested in ethical theory and

the more the theoretical discussion has become distanced from argument about

practical affairs.15 Anyone inexperienced in ethical theory will find it difficult to

follow the analysis, to grasp its methodological importance, or to see its practical

importance.

To the extent that Daniels has been successful in bioethics—where his con-

tribution to issues of distributive justice is unmistakable—his success seems to me

curiously detached from his theory of justification. He asserts that throughout his


bioethical writings he is appealing to and employing the method of reflective

equilibrium. Perhaps he is, but it is obscure what the method requires, how one

could check on whether the method has been used, and even whether Daniels

follows his model of justification. I am not asserting that Daniels fails to follow

his method. I am asserting that it has never been made clear how the method

connects to practical problems, how one would know whether it has been

followed, and how it might be used by others in bioethics.16 Daniels states

that “the best defense” of the method of reflective equilibrium is “to put it

into action . . . and let it be judged by its results.”17 Fair enough, but to assess the

results, one must understand what one is assessing. The theory is elegant; the

bearing on practice remains more promise than achievement. It continues to be

unclear whether anyone in bioethics has followed the reflective-equilibrium

model (I include myself), despite its standing as the most widely mentioned

model.18

Substantial contributions in bioethics to the theory and practice of justification

are rare. One can cite examples such as Bernard Gert’s able justification of the

rules in his moral system,19 but, like most of Daniels’s work, it is not easy to see

how this philosophical accomplishment is connected to practical problems of

bioethics.

Practice Standards as the Model

Daniels aspires to a close link between theory and practice, but some moral

philosophers in bioethics imply that practice has no close confederation with

theory. These writers seem to recoil from the abstractness of theory and its

distance from practice.

Few theoretical discussions have been more influential in recent bioethics than

Alasdair MacIntyre’s conception of a moral practice as a cooperative arrangement

in pursuit of goods that are internal to a structured communal life.20 Building on

this conception, various philosophers have proposed that what it means to be a

good practitioner in health care and research rests on internal (and sometimes

evolving) standards of practice, rather than on an external moral theory. Here is a

statement by Howard Brody and Frank Miller:

Physicians, by virtue of becoming socialized into the medical profession,

accept allegiance to a set of moral values which define the core nature of

medical practice. These values give rise to at least some of the moral duties

incumbent upon physicians in their professional role and indicate the

virtues proper to physicians. . . . The professional integrity of physicians is

constituted by allegiance to this internal morality.21


Brody and Miller may seem to be illicitly grounding duties and virtues in

existing practice standards without justifying those standards. However, they

note that “Even the core of medical morality must be thoughtfully reevaluated

and reconstructed at intervals.”22 They observe that existing standards of practice

may be shallow and in need of reexamination. Their test case of standards of

practice currently in need of reevaluation is medical morality’s prohibition of

physician-assisted suicide. Although medical tradition has insistently condemned

this activity, Brody and Miller say, “today’s physicians might [legitimately] still

conclude that so many things have changed since the time of Hippocrates . . . [that

the times] warrant a reconstruction of the internal morality so that assisted

suicide in certain defined circumstances is permissible.”23

What would “warrant a reconstruction”? Brody and Miller propose that phy-

sician-assisted death may turn out, upon careful consideration, to be justifiable in

terms of the internal morality of medicine. The first line of justification, they

suggest, should be to show that internal standards of professional integrity and

moral duty are not violated by active assistance in dying. However, realizing that

this morality may not support the entirety of needed reforms, Brody and Miller

propose that external standards will sometimes be needed:

Professional integrity does not encompass the whole of medical ethics.

Moral considerations other than the norms of professional integrity may

be appealed to in favor of, or against, permitting a practice of limited

physician-assisted death. . . . An experimental public policy of legalizing

the practice should be undertaken, subject to stringent regulatory

safeguards.24

What sort of “moral considerations” would supply a justification? One might

reasonably expect moral philosophers such as Brody and Miller to turn to moral

philosophy. Having asserted that justification is a central issue and that internal

justification using practice standards may come up short, they make no attempt at

justification beyond a vague statement that “the reconstruction will be carried out

by those who live in modern society who are inevitably influenced by societal

values as they interpret the history.”25 In short, their first-line appeal is to internal

standards of medical morality; their second-line appeal is to societal values; and

there is no third-line appeal to ethical theory or even a hint that it might be

relevant.

I am not arguing that Brody and Miller are mistaken. I am pointing out that

their training in ethical theory is removed from their venture and that their

approach is characteristic of much of the current literature by philosophers in

bioethics. Many prefer to ground their claims in institutional standards, societal

values, paradigm cases, narratives—indeed, in virtually any source other than


ethical theory. When, subsequently, it appears that neither institutions nor

societies provide the needed justification, few hints are provided about how to

proceed. Just at the point ethical theory seems relevant, it is ignored. This neglect

transparently conveys the impression that moral philosophy is ill-equipped to

help us solve problems.

CONCEPTUAL ANALYSIS

I turn now to the third of the three areas I said I would investigate, conceptual

analysis. This method is of the first importance in moral philosophy. Although it

is rarely expected to resolve practical moral problems, it is widely thought to play

an important role in such argument.

One example in bioethics of the philosophical and practical importance of

conceptual analysis is work on the concept of “informed consent.” Appropriate

criteria must be identified if we are to define and classify an act as an informed

consent. Overdemanding criteria such as “full disclosure and complete under-

standing” mischaracterize informed consent, and so-called informed consents are

then impossible to obtain. Conversely, underdemanding criteria such as “the

patient signed the form” misconceive informed consent, and so-called informed

consents then become too easy to obtain. In the literature of informed consent

over the last 30 years, conceptual and moral analysis of informed consent has,

beyond a reasonable doubt, been vital to advances in both clinical and research

ethics. This work is ongoing and incomplete, but it has changed our under-

standing as well as our practices in research and clinical settings.

The Concept of Autonomy

Though the notion of informed consent has received extensive attention in

bioethics, many other concepts have not received comparable attention—or at

least not with an eye to professional and policy areas. Paradoxically, the more

fundamental and philosophically challenging the concept, the less it seems that

moral philosophers in bioethics have attended to it or spelled out a theory’s

implications for practice. I will use “autonomy” as an example. In moving to this

example, I start with an engaging 1988 statement by Jonathan Moreno that was

intended to motivate his important research on the role of consensus in

bioethics: “Unlike the concept of autonomy, consensus has not yet undergone

systematic and intense scrutiny in medical ethics.”26 I agree about consensus,

but still today no clear connection has beenmade between conceptual analysis of

autonomy and practical areas of bioethics.


It is agreed in the literature that “autonomy,” “respect for autonomy,” and

“rights of autonomy” are very different concepts. “Respect for autonomy” and

“rights of autonomy” are moral notions, whereas “autonomy” and “autonomous

person” are not obviously moral notions. To many writers the latter seem more

metaphysical than moral, and literature in both ethical theory and bioethics

generally supports this idea.27 However, the distinction between the metaphysical

concept of autonomy and the moral concept of autonomy has fostered confusing

views, and uncertainty currently surrounds the meaning of autonomy, its rela-

tionship to the concept of persons, the descriptive or normative character of these

concepts, and the connection of these notions to that of respect for autonomy

(and respect for persons). What it is about autonomy that we are to respect

remains unclear, and it remains obscure what “respect” means. Most obscure of

all is how practice is affected by a theory of autonomy.

The contemporary literature in bioethics contains no theory of autonomy that

spells out its nature, its moral implications, its limits, how respect for autonomy

differs from respect for persons (if it does), and the like. Much of the literature

relies on notions of liberty and agency. Themost detailed theory of autonomy by a

writer in bioethics is Gerald Dworkin’s.28 He offers a “content-free” definition of

autonomy as a “second-order capacity of persons to reflect critically upon their

first-order preferences, desires, wishes, and so forth and the capacity to accept or

attempt to change these in the light of higher-order preferences and values.”29 For

example, consider a person who has desires for low-quality foods (a first-order

desire or preference). To qualify as autonomous, the person must have the

capacity to reflect on his or her desires and form a desire about whether to

change or retain those desires (a second-order desire or preference).

This theory has many problems. There is nothing to prevent a reflective

acceptance or preference at the second level from being caused by and assured

by the strength of a first-order desire. The individual’s acceptance of or identifica-

tion with the first-order desire would then be no more than a causal result of an

already formed structure of preferences, not a new structuring of preferences. For

example, if I decide that I prefer to continue eating low-quality food because I

have a passion for doing so, this choice seems determined by my initial desires. At

first I wanted cheeseburgers; then I discovered that I desire to desire cheeseburgers

and that I want this second-order desire to be my controlling desire. This example

may seem superficial, but the two-tier theory embraced by Dworkin and

Frankfurt lacks adequate conditions to protect even against such a simple case.

To make the theory plausible, a component, supplementary theory would have to

be added that distinguished between influences or desires that rob an individual of

autonomy and influences or desires consistent with autonomy.30

More worrisome for present purposes is that, like Daniels’s account of

justification, Dworkin’s theory of autonomy has not been carefully assessed or


used in bioethics. Its implications for practice are unclear and untested. This

neglect is not surprising, as there seem to be no practical implications of the

theory. Dworkin himself does not appear to use the theory in his writings on

applied topics of bioethics. He devotes the second half of his book The Theory

and Practice of Autonomy to “Practice,” where he treats five issues in bioethics.

He scarcely mentions the theory developed in the first half of the book and never

attempts to decide an important question in bioethics by reference to the theory.

In his “Epilogue” to the book, Dworkin points out that a number of conceptual

issues must be worked out to develop the research program he initiated. It is two

decades later, and nothing has appeared.

If I could point to a more penetrating or successful theory of autonomy in

bioethics than Dworkin’s, I would, but it does not exist. We await a carefully

crafted philosophical theory as well as a connection between theory and practice.

The Concepts of Killing and Letting Die

Consider, as a second example of skimpy conceptual analysis, the distinction

between killing and letting die. Many issues have been raised in the literature of

bioethics about the conceptual conditions under which physicians do or do not

kill patients, let them die, or neither. Most recently, physician-assisted suicide has

been the primary concern. Various problems of moral justification are at work in

these debates, but I will focus here entirely on the concepts that drive the

discussion: killing and letting die.

In medical tradition killing by physicians is prohibited and letting die per-

mitted under specified conditions. Despite a remarkable convergence of opinion

in support of this traditional outlook, no one has yet produced a cogent analysis

of the distinction between killing and letting die as it functions in medicine so

that meaningful law and professional ethics can be developed through the

distinction. Philosophical writers on the subjects of euthanasia, assisted suicide,

terminal sedation, and the like are typically concerned about killing rather than

letting die. However, their work does little to explicate the conditions under

which killing by physicians occurs; indeed, the work is sketchy to nonexistent on

whether “physician-assisted suicide” actually involves suicide by patients. I am

not aware of a philosophical article or book on killing that carefully addresses

what counts as killing in medical contexts and how that notion is to be

distinguished from letting die.31

The notion of letting patients die has often been ignored in this literature,

though conceptually it is no less important than killing. “Letting die” occurs in

medicine under two circumstances: cessation of medical technology because it is

useless and cessation of medical technology because it has been validly refused.


The latter notion applies to both first-party and proxy refusals, though I will not

address this distinction here. Honoring a valid refusal of a useful treatment

knowing of a fatal outcome is a letting die, not a killing. The type of action—a

killing or a letting die—can depend entirely on whether a valid refusal justifies

the forgoing of medical technology. This brings evaluative appraisal into the

heart of the concept. (See my Essay 7 on hastened death in these Collected

Essays.)

In the medical context, “letting die” is value laden in the sense that it is tied to

acceptable acts, where acceptability derives either from the futility of treatment or

a valid refusal. Killing, by contrast, is conceptually tied in medical contexts to

unacceptable acts. Thus, the value neutrality of “killing” and “letting die” found in

ordinary moral discourse (as when we say “the disease killed her”; “she was killed

in an auto accident”; or “he killed an enemy soldier”) is not present in medical

morality, where letting die is prima facie justified and killing prima facie unjus-

tified. A thorough philosophical analysis would have to work through the subtle-

ties of these problems. This is the sort of problem that philosophers should be well

equipped to address and also the sort of problem that stands to affect medical

practice and contribute to bioethics.

However, philosophers generally miss these key issues. A common strategy in

philosophical literature is to present two cases differing only in that one is a

killing and the other a letting die (usually involving some harm or wrong), and

then to investigate conceptually what the difference is and morally whether the

difference is relevant to judgments of right and wrong. The hypothesis is that the

conceptual or moral difference in the two cases shows that the distinction either

is or is not viable or morally relevant. In treating such cases, philosophers seem

to prefer to retreat from the messiness of the medical situation—that is, they

distance their analyses from the real problems of practice—to achieve concep-

tual clarity and deal with problems of concern to fellow philosophers. They

utilize cases of runaway trolleys, minuscule releases of toxic substances, persons

who drown in bathtubs, rescuing drowning persons by killing someone in the

path of the swimmer, and the like. These examples and accompanying analyses

fail to illuminate whether doctors kill or let die, whether they act justifiably or

not, and how to think about the acts of patients who refuse life-saving

technologies.

In short, philosophical literature tends to evade the central context in which the

distinction arises in bioethics—real encounters between doctors and patients—

and, as a result, the analyses turn out to be too pure and general to illuminate that

context. The fit is poor. I do not say that philosophical analysis fails to illuminate

circumstances of killing or letting die beyond medical practice. I claim only that

the available conceptual analyses are largely unilluminating in regard to medical

practice, which is the domain of bioethics.


CONCLUSION

The controversies I have discussed largely originated in philosophy and have

been perpetuated using philosophical discourse. Consequently, outsiders have

had a hard time looking in. The more the issues have been philosophically

refined, the less serviceable and influential ethical theory has become in

bioethics. Philosophical theories today have a diminished stature in bioethics

by comparison to their standing of 10 or 20 years ago. The reasons for the

demotion of ethical theory are the lack of distinctive authority behind any one

framework or methodology, the unappealing and formidable character of many

theories, the indeterminate nature of general norms of all sorts, the turn in

bioethics to more practical issues, and—most important—the stumbling and

confusing manner in which philosophers have attempted to link theory to

practice.

How theory can be connected to practice is a problem of greater urgency today

than it was 30 years ago, when philosophers were not in touch with medical

morality and virtually nothing was expected of them in this domain. In 1974

Samuel Gorovitz and I drove back and forth from Washington, D.C., to

Haverford College over a six-week period to discuss with a group of philosophers

whether bioethics had a future in philosophy. The word bioethics had just been

coined, and no one understood quite what it meant or encompassed. Few

philosophers had exhibited an interest in the subject. While traveling on

Interstate 95, Gorovitz and I reflected on how developments in biomedicine

present a treasure chest of fundamentally philosophical issues. Gorovitz was

puzzled that philosophers ignored such matters. He thought these issues “a

natural” for philosophers. I think he was right, and I still believe that many

issues in bioethics are fundamentally problems that should be handled by the

methods of moral philosophy. However, thousands of publications later, the

grand promise of ethical theory for bioethics remains mostly promise.

For 15 years or so it seemed that Gorovitz would be proved right: Bioethics did

seem a natural for philosophers. In limited respects he has been proven correct

by the profound influence that philosophers have enjoyed in various areas of

bioethics. However, the literature of an interdisciplinary bioethics has shifted in

the last two decades or so toward greater levels of comfort with law, policy

studies, empirical studies, standards of practice, government guidelines, and

international guidelines—progressively marginalizing the role of ethical theory.

Moral philosophers have not convinced the interdisciplinary audience in

bioethics, or themselves, that ethical theory is foundational to the field and

relevant to practice. To continue on this path will call into question whether

ethical theory has a significant role in bioethics. It is a question worth

considering.


Notes

1. See R. K. Fullinwider, “Against Theory, or: Applied Philosophy-A Cautionary Tale,”

Metaphilosophy 20 (1989): 222–234.

2. For works published during the period, see National Commission for the Protection of

HumanSubjects,TheBelmontReport: EthicalGuidelines for the ProtectionofHumanSubjects of

Research (DHEW Publication No. (OS) 78-0012, 1978); J. V. Brady and A. R. Jonsen, “The

Evolution of Regulatory Influences onResearchwithHumanSubjects,” inR.Greenwald, et al.

eds.,Human Subjects Research (New York: Plenum Press, 1982): 3–18; R. J. Levine, Ethics and

Regulation of Clinical Research, 1st edn. (Baltimore: Urban & Schwarzenberg, 1981); E. D.

Pellegrino andD. C. Thomasma,A Philosophical Basis of Medical Practice (New York: Oxford

University Press, 1981); andF.W.O’Connor, et al.Deviance andDecency:TheEthics ofResearch

With Human Subjects (London: Sage Publications 1979). For later studies or accounts of the

period, see J. D. Moreno, Deciding Together: Bioethics and Consensus (New York: Oxford

University Press, 1995); Advisory Committee onHumanRadiation Experiments, Final Report

of the Advisory Committee on Human Radiation Experiments (New York: Oxford University

Press, 1996); D. DeGrazia, “Moving Forward in Bioethical Theory: Theories, Cases, and

Specified Principlism,” Journal of Medicine and Philosophy 17 (1992): 511–539; J. Arras,

“Principles and Particularity: The Role of Cases in Bioethics,” Indiana Law Journal 69

(1994): 983–1014; and R. M. Veatch, “From Nuremberg through the 1990s: The Priority of

Autonomy,” inH. Y. Vanderpool, ed.The Ethics of Research Involving Human Subjects: Facing

the 21st Century (Frederick, MD: University Publishing Group 1996): 45–58.

3. K. D. Clouser and B. Gert, “A Critique of Principlism,” The Journal of Medicine and

Philosophy 15 (1990): 219–236. A. Jonsen and S. Toulmin, The Abuse of Casuistry (Berkeley:

University of California Press, 1988).

4. N. A. Christakis, “Ethics are Local: Engaging Cross-Cultural Variation in the Ethics

for Clinical Research,” Social Science and Medicine 35 (1992): 1079–1091, quotation from

1079, 1089.

5. Ibid., 1088.

6. L. Turner, “Zones of Consensus and Zones of Conflict: Questioning the ‘Common

Morality’ Presumption in Bioethics,” Kennedy Institute of Ethics Journal 13 (2003): 193–218,

esp. 196.

7. Christakis, supra note 4, 1088.

8. R. Macklin, Against Relativism: Cultural Diversity and the Search for Ethical Universals

in Medicine (New York: Oxford University Press, 1999).

9. D. P. Sulmasy, “Review of Against Relativism,” The National Catholic Bioethics

Quarterly 1 (2001): 467–469, esp. 468–469. See, however, Macklin’s explicit statement in

her “Preface” (p. v.) regarding the way she limits her theoretical goals.

10. For an example of the sorts of arguments of the opposition in publications in bioethics

that need to be addressed, see the essays inCross-Cultural Perspectives on the (Im)Possibility of

Global Bioethics, ed. Julia Tao Lai Po-Wah (Dordrecht: Kluwer, 2002).

11. For nuanced accounts of the use of moral philosophy to defend both relativism and

antirelativism, see G. Harman and J. J. Thomson, Moral Relativism and Moral Objectivity


(Cambridge, MA: Blackwell Publishers, 1996). Thomson mounts a compelling philosophical

defense of “objectivity.” A very different and engaging approach, more concrete but not

appropriately described as practical ethics, is provided by M. C. Nussbaum, Women and

Human Development: The Capabilities Approach (Cambridge: Cambridge University Press,

2000), especially chapter 1, “In Defense of Universal Values.”

12. N. Daniels, Justice and Justification. Reflective Equilibrium in Theory and Practice

(Cambridge: Cambridge University Press, 1996); N. Daniels, “Wide Reflective Equilibrium

in Practice,” in L. W. Sumner and J. Boyle, eds., Philosophical Perspectives on Bioethics

(Toronto: University of Toronto Press, 1996), 96–114; N. Daniels, “Wide Reflective

Equilibrium and Theory Acceptance in Ethics,” Journal of Philosophy 76 (1979), 256–282;

and, with A. Buchanan, D. W. Brock, and D. Wikler, From Chance to Choice: Genetics and

Justice (Cambridge: Cambridge University Press, 2000), esp. 371ff.

13. J. Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971; revised

edition, 1999).

14. Daniels, supra note 12 (1).

15. N. Daniels, “Wide Reflective Equilibrium in Practice” provides the most accessible

material on methodology in bioethics, but the claim to have used the method in practice is

boldest andmost interesting inFromChance toChoice:Genetics and Justice, 371; and also 22–23,

308–309.

16. In Justice and Justification: Reflective Equilibrium in Theory and Practice, Daniels

discusses reflective equilibrium, especially wide reflective equilibrium, in the first part of

the book—the “Theory” part—but not in the second. Chapters 1–8 in the first part deal

with the theory of reflective equilibrium. In the second part (chapters 9–16)—the

“Practice” part—reflective equilibrium is mentioned only in the title of chapter 16, and

never developed. In chapter 9, the first of the practice chapters, Daniels points to some

general implications of his account and rightly recognizes “the many difficulties that face

drawing implications from ideal theory for nonideal settings” (199). As I readDaniels, he is

a persuasive spokesperson for the defense and extension of Rawls’s theory of justice, but he

does little to implement transparently the method of reflective equilibrium or to show its

implications for practical ethics.

17. N. Daniels, From Chance to Choice: Genetics and Justice, supra note 12, 371.

18. The book Reflective Equilibrium, ed. Webren van der Burg and Theo van

Willigenburg (Dordrecht: Kluwer Academic Publishers, 1998), is composed of 19 essays

on the subject. It supposedly discusses how the idea of reflective equilibrium offers a model

for practical moral problems. While there is philosophical discussion of the practical

implications of the model, I cannot find actual examples of the use of the method to

address practical problems.

19. B. Gert, Morality: A New Justification of the Moral Rules. This theory is brought to

bear on various problems in bioethics in B. Gert, C. M. Culver, and K. D. Clouser, Bioethics:

A Return to Fundamentals (New York: Oxford University Press, 1997).

20. A. MacIntyre, After Virtue, 2nd edn. (Notre Dame: University of Notre Dame Press,

1984), 17, 175, 187, 190–203. See also A.MacIntyre, “Does Applied Ethics Rest on AMistake?”

Monist 67 (1984): 498–513.


21. H. Brody and F. G. Miller, “The Internal Morality of Medicine: Explication and

Application to Managed Care,” Journal of Medicine and Philosophy 23 (1998): 384–410,

especially 386.

22. Brody and Miller, supra note 21, 393–397; italics added.

23. Brody and Miller, supra note 21, 397.

24. F. G. Miller & H. Brody, “Professional Integrity and Physician-Assisted Death,”

Hastings Center Report (1995): 8–17, especially 12–16.

25. Brody and Miller, supra note 21, 393–394, 397; italics added.

26. J. Moreno, “Ethics by Committee: The Moral Authority of Consensus,” The Journal

of Medicine and Philosophy 13 (1988), 413.

27. D. Dennett, “Conditions of Personhood,” in The Identities of Persons, ed. A. O. Rorty

(Berkeley: University of California Press, 1976), 175–196; J. Feinberg and B. Baum

Levenbook, “Abortion,” in Matters of Life and Death: New Introductory Essays in Moral

Philosophy, 3rd edn., ed. T. Regan (New York: Random House, 1993); M. A. Warren,

“Abortion,” in A Companion to Ethics, Peter Singer, ed. (Cambridge, MA: Blackwell

Reference, 1991); 303–314; M. A. Warren, Moral Status: Obligations to Persons and Other

Living Things (New York: Oxford University Press, 1997); and Kennedy Institute of Ethics

Journal 9 (December 1999), Special Issue on Persons, ed. Gerhold K. Becker.

28. G. Dworkin, The Theory and Practice of Autonomy (New York: Cambridge University

Press, 1988). The most interesting book in bioethics on autonomy, in my judgment, is

O. O’Neill, Autonomy and Trust in Bioethics (Cambridge: Cambridge University Press,

2002); however, this book does not present a theory or conceptual analysis of autonomy.

29. Dworkin, supra note 28, 20.

30. No such work has as yet been put forward in bioethics, but some beginnings are

available in ethical theory. See H. G. Frankfurt, “Freedom of the Will and the Concept of a

Person,” Journal of Philosophy 68 (1971): 5–20, as reprinted in The Importance of What We

Care About (Cambridge: Cambridge University Press, 1988): 11–25; and J. Christman,

“Autonomy and Personal History,” Canadian Journal of Philosophy 21 (1991): 1–24, esp.

10ff. Christman’s account is a nuanced version of the theory of nonrepudiated acceptance

that uses the language of nonresistance. Although it is not clear that Frankfurt has a theory

of autonomy, see his uses of the language of “autonomy” in hisNecessity, Volition, and Love

(Cambridge: Cambridge University Press, 1999), chaps. 9, 11. Frankfurt’s early work was on

persons and freedom of the will. In his later work, he seems to regard the earlier work as

providing an account of autonomy, which is a reasonable estimate even if it involves some

creative reconstruction.

31. The articles or essays closest to this ideal are J. L. Bernat, B. Gert, and R. P.

Mogielnicki, “Patient Refusal of Hydration and Nutrition: An Alternative to Physician-

Assisted Suicide or Voluntary Active Euthanasia,” Archives of Internal Medicine 153

(December 27, 1993): 2723–2728; D. W. Brock, Life and Death: Philosophical Essays in

Biomedical Ethics (New York: Cambridge University Press, 1993); T. L. Beauchamp, “The

Justification of Physician-Assisted Suicide,” Indiana Law Review 29 (1996): 1173–1200; and

J. McMahan, “Killing, Letting Die, and Withdrawing Aid,” Ethics 103 (1993): 250–279.


14

THE FAILURE OF THEORIES OF PERSONHOOD

What it is to be a person is a principal topic of metaphysics. Ideally, a metaphysical

theory expresses a morally detached interest in how to distinguish persons from

nonpersons. However, the metaphysics of persons has often been put to work to

defend a preferred moral outcome, placing metaphysics in the service of ethics.

Metaphysics is invoked to inquire whether individuals have rights and whether

the theory of persons can address practical problems of abortion, reproductive

technology, infanticide, refusal of treatment, senile dementia, euthanasia, the

definition of death, and experimentation on animals.

The different objectives of theories of persons can be clarified by a distinction

between metaphysical and moral concepts of persons.1 Metaphysical person-

hood is composed entirely of a set of person-distinguishing psychological

properties such as intentionality, self-consciousness, free will, language acquisi-

tion, pain reception, and emotion. The metaphysical goal is to identify a set of

psychological properties possessed by all and only persons. Moral personhood,

by contrast, refers to individuals who possess properties or capacities such as

moral agency and moral motivation. These properties or capacities distinguish

moral persons from all nonmoral entities. In principle, an entity could satisfy all

the properties requisite for metaphysical personhood and lack all the properties

requisite for moral personhood.

However, most published theories of persons are not clearly distinguishable

into these types or even attentive to the distinction between metaphysical and

moral personhood. Proponents of these theories have generally not even

approached the subject using these distinctions. Their goal has primarily

been to delineate the distinctive properties of personhood—moral or non-

moral—that are necessary for and confer moral status on an individual. For

three decades, and arguably for several centuries,2 the dominant trend in the

literature on persons has been to delineate nonmoral, principally cognitive,

properties of individuals in a metaphysical account, from which conclusions

can be drawn about moral status. A good example is Michael Tooley’s well-

known analysis moving from metaphysical premises to moral conclusions

(1972, sec. 3):

What properties must something have in order to be a person, i.e., to have a

serious right to life? The claim I wish to defend is this: An organism possesses

a serious right to life only if it possesses the concept of a self as a continuing

subject of experiences and other mental states, and believes that it is itself a

continuing entity.

Tooley (1983, p. 51; cf. p. 35) observes that

[I]t seems advisable to treat the term “person” as a purely descriptive term,

rather than as one whose definition involves moral concepts. For this appears

to be the way the term “person” is ordinarily construed.3

In this account, “person” has descriptive content (“an entity that possesses

either self-consciousness or rationality”), but is not a purely descriptive term; the

person-making properties in this metaphysical account endow their possessors

with moral rights or other moral protections.

The belief persists in philosophy and beyond that some special cognitive

property or properties of persons such as self-consciousness or rationality

confers a unique moral status, and perhaps forms the exclusive basis of this

status.4 I believe, by contrast, that no cognitive property or set of such properties

confers moral status and that metaphysical personhood is not sufficient for

either moral personhood or moral status, though some conditions of metaphy-

sical personhood may be necessary conditions of moral personhood.5 I also

maintain that moral personhood is not a necessary condition of moral status.

I claim, in this way, that metaphysical personhood does not entail either moral

personhood or moral status and that personhood of either type is not the only

basis of moral status.


THE CONCEPT OF METAPHYSICAL PERSONHOOD

The common-sense concept of person is, roughly speaking, identical with the

concept of human being. Human psychological properties continue to play a

pivotal role in philosophical controversies over personhood. However, there is no

warrant for the assumption that only properties distinctive of membership in the

human species count toward personhood or confer moral status. Even if certain

properties that are strongly correlated with membership in the human species

qualify humans more readily than the members of other species, these properties

are only contingently connected to being human. It just so happens, if it is so at all,

that individuals possessing these properties are of a particular natural species. The

properties could be possessed by members of nonhuman species or by entities

outside the sphere of natural species such as computers, robots, and genetically

manipulated species.6

Fortunately, a metaphysical account of persons need have no reference to

properties possessed only by humans. In some of the theories mentioned pre-

viously (e.g., Tooley’s), an entity is a person if and only if it possesses certain

cognitive rather than singularly human properties. Cognitive conditions of meta-

physical personhood similar to the following have been promoted by several

classical and contemporary writers7: (1) self-consciousness (of oneself as existing

over time), (2) capacity to act on reasons, (3) capacity to communicate with others

by command of a language, (4) capacity to act freely, and (5) rationality.

Presumably these characteristics are put forward to distinguish persons from

nonpersons irrespective of species, origin, or type. If so, it is an open question

whether a robot, a computer, an ape, or God qualifies for metaphysical person-

hood. Methodologically, the properties of personhood are presumed to be deter-

minable a priori by consulting our shared concept of person; such a theory does

not require empirical discovery. The only empirical question is whether an entity

in fact satisfies the conceptual conditions. A classical example of this method is

found in John Locke’s analysis of a person as a “thinking intelligent being, that has

reason and reflection, and can consider itself as itself, the same thinking thing in

different times and places” (1975, 2.27.9; see also 2.27.24–26). Locke pointed out

that despite the close association between “man” and “person,” the two concepts

are distinct, a claim he defended by presenting cases to show that the same man

need not be the same person.

Sometimes it is said by those who defend criteria resembling (1) through (5)

that only one of these criteria must be satisfied for metaphysical personhood—for

example, self-consciousness, rationality, or linguistic capacity. Other writers

suggest that each condition must be satisfied and that the five conditions are

jointly necessary and sufficient. Another view is that some subset of these five

conditions is both necessary and sufficient.

T H E F A I L U R E O F T H E O R I E S O F P E R S O N H O O D 249

PROBLEMS IN THEORIES OF METAPHYSICAL PERSONHOOD

These cognitive theories fail to capture the depth of commitments embedded in

using the language of “person,” and they sometimes confuse the most important

issues by moving from a purely metaphysical claim to a claim about either moral

personhood or moral status. By themselves these cognitive properties have no

moral implications. Such implications occur only if an analysis assumes or

incorporates an independent moral principle, such as “respect for persons.”

Such a principle, being independent of a metaphysical theory, would have to be

defended independently and given suitable content.

Suppose that some being is rational, acts purposively, and is self-conscious.

How is moral personhood or any form of moral status established by this fact? Do

moral conclusions follow from the presence of these properties? An entity of this

description need not be capable of moral agency or able to differentiate right from

wrong; it may lack moral motives and all sense of accountability. It may perform

no actions that we can judge morally. It might be a computer, a dangerous

predator, or an evil demon. No matter how elevated our respect for this entity’s

cognitive capacities, these capacities will not amount to moral personhood and

will not establish any form of moral status. Capacities of language, rationality,

self-consciousness, and the like lack an intrinsic connection to moral properties

such as moral agency and moral motivation.

A property often cited in the metaphysical hunt, as we saw in Tooley’s theory, is

self-consciousness—that is, a conception of oneself as persisting through time and

having a past and a future. If animals such as birds and bears lack self-awareness

and a sense of continuity over time, they lack personhood (see, e.g., Buchanan &

Brock, 1989, pp. 197–199; Dworkin, 1988, esp. Chapter 1; Harris, 1985, pp. 9–10).

The hypothesis is that although animals exhibit goal-directed behaviors such as

building a nest, they do so without any sense of self, which is thought to be

essential to personhood or to any condition that confers moral status. However, it

is more assumed than demonstrated in these theories that nonhuman animals

lack a relevant form of self-consciousness or its functional equivalent. The prima

facie evidence of various types and degrees of animal self-awareness is striking and

the possibility of self-consciousness cannot be dismissed without careful study.

Language-trained apes appear to make self-references, and many animals learn

from the past and then use their knowledge to forge plans of action for hunting,

stocking reserve foods, and constructing dwellings (see Griffin, 1992). These

animals are aware of their bodies and their interests, and they unerringly distin-

guish those bodies and interests from the bodies and interests of others. In play

and social life, they understand assigned functions and decide for themselves what

roles to play. A few appear to recognize themselves from reflections in mirrors

(cf. DeGrazia 1997, p. 302; Gallup, 1977; Miles, 1993; Patterson & Gordon, 1993).


There may, then, be reason to attribute at least elementary self-consciousness to

these animals, and to think of this ability as admitting of degrees in the several

criteria that might be used to analyze it.

One possible strategy to avoid this conclusion is to increase the demands built

into the concept of self-consciousness. Harry Frankfurt’s (1971) account, some-

times presented as a theory of autonomy, could be adapted to this end (see also

Dworkin, 1988, Chapters 1–4; Ekstrom, 1993). In this theory, all and only persons

have a form of self-consciousness involving distanced self-reflection. Persons

reflectively judge and identify with their basic, first-order desires through

second-order desires, judgment, and volition. Second-order mental states have

first-order mental states as their intentional objects, and considered preferences

are formed about first-order desires and beliefs. For example, a long-distance

runner may have a first-order desire to run several hours a day, but also may have

a higher-order desire to decrease the hours and the level of commitment. Action

from the second-order desire is autonomous and is characteristic of the person;

action from the first-order desire is not autonomous and is typical of animal

behavior. The capacity to rationally accept or repudiate lower-order desires or

preferences, which involves a lofty cognitive ability of distanced self-reflection, is

the centerpiece of this account.

Several problems haunt this theory. First, there is nothing to prevent a reflective

acceptance or repudiation at the second level from being caused by and assured by

the strength of a first-order desire. The individual’s acceptance of or identification

with the first-order desire would then be no more than a causal result of the

already formed structure of preferences, not a new structuring of preferences, and

so not a particularly attractive criterion of personhood. Second-order desires

would not be significantly different from or causally independent of first-order

desires, other than being second-order. Second, the conditions of distance and

reflective control are so demanding in this theory that either many human actors

will be excluded as persons or their actions will be judged nonautonomous. It is

doubtful that an identification at the second level is present in most of the actions

that we commonly perform. A potential moral price of this demanding theory is

that individuals who have not reflected on their desires and preferences at a higher

level deserve no respect for actions that derive from their most deep-seated desires

and preferences.8

As the quality or level of required cognitive activity is reduced in a theory to

accommodate these problems, the volume of humans who qualify will increase,

but so will the volume of nonhuman animals. Less demanding conditions—

understanding and self-control, say—will likely be satisfied to a greater or lesser

extent. A threshold line will have to be drawn in a theory that separates an

adequate degree of understanding and self-control from an inadequate degree.

Again, a high threshold will exclude many humans that we normally regard as


autonomous persons; a low threshold will include at least some nonhuman

animals. Nonhuman animals are especially likely to qualify in a theory that uses

low cognitive criteria by contrast to lofty cognitive criteria.9

Virtually all criteria of personhood or autonomy admit of degrees, and most

develop over time. Rationality and understanding clearly admit of degrees (though

self-consciousness is a more challenging case). A theory that embraces such degrees

of autonomy—and perhaps thereby degrees of personhood—must allow for the

possibility that some nonhuman animals will be positioned at a higher level of

autonomy or personhood than some humans.10 The fact that humans will generally

score higher under these criteria than other species of animals is a contingent fact.

A nonhuman animal may overtake a human whenever the human loses a sufficient

measure of cognitive abilities. If, for example, the primate in training in a language

laboratory exceeds the deteriorating Alzheimer’s patient on the relevant scale of

cognitive capacities, the primate gains a higher degree of personhood, andmay gain

a higher moral status. However, metaphysical theories of persons that appeal

exclusively to cognitive criteria entail no such conclusions about either moral

persons or moral status, two topics to which I now turn.

THE CONCEPT OF MORAL PERSONHOOD

By comparison to metaphysical personhood, moral personhood is relatively

uncomplicated. I will not attempt an account of the necessary and sufficient

conditions of moral personhood, but it seems safe to assume that a creature is a

moral person if (1) it is capable of making moral judgments about the rightness

and wrongness of actions and (2) it has motives that can be judged morally. These

are two jointly sufficient conditions of personhood. They are moral-capacity

criteria and also cognitive criteria, but they are not sufficient conditions of

morally correct action or character; an individual could be immoral and still

qualify for moral personhood. These criteria will require for their explication

some cognitive conditions discussed previously. For example, the capacity to

make moral judgments may require rationality. A general theory of moral person-

hood, then, would be needed to defend the aforementioned two conditions and to

relate them to the cognitive conditions discussed previously.

However, such a general theory is not needed for the two primary theses that

I now will defend. The first thesis is that moral personhood, unlike cognitivist

theories of metaphysical personhood, is sufficient for moral status. Moral agents

are paradigm bearers of moral status. Any entity qualifying for moral personhood

is a member of the moral community and qualifies for its benefits, burdens,

protections, and punishments. Moral persons understand moral reciprocity and

the communal expectation that they will treat others as moral persons. It is central


to the institution of morality itself that moral persons deserve respect and are to be

judged as moral agents. Moral persons know that we can condemn their motives

and actions, blame them for irresponsible actions, and punish them for immoral

behavior. The moral protections afforded by this community may be extended to

the weak and vulnerable who fail to qualify as moral persons, but moral status for

these individuals must rest on some basis other than moral personhood.

The second thesis is that nonhuman animals are not plausible candidates for

moral personhood, though the great apes, dolphins, and perhaps other animals

with similar properties could turn out to be exceptions.11 Here I borrow from

Charles Darwin (1981, Chapter 3). He denies that animals make moral judgments

while affirming that they sometimes display moral emotions and dispositions. For

example, he maintains that animals do not make genuine judgments of blame

when they punish their peers for misbehavior, but that they do display love,

affection, and generosity. Darwin described conscience (the moral sense in

humans) as “the most noble of all the attributes” found in the human animal:

“I fully subscribe to the judgment of those writers who maintain that of all the

differences between man and the lower animals, the moral sense or conscience is

by far the most important.” Darwin thus regarded nonhuman animals as failing

the test of moral personhood.12

Humans, too, fail to qualify as moral persons if they lack one or more of the

conditions of moral personhood. If moral personhood were the sole basis of

moral rights (a view I do not hold), then these humans would lack rights—and

precisely for the reasons that nonhuman animals would. Unprotected humans

would presumably include fetuses, newborns, psychopaths, severely brain-

damaged patients, and various dementia patients. I will now argue that these

individuals do possess some rights and merit moral protections, but not because

they satisfy the conditions of moral personhood. In this respect, these humans are

in the same situation as many nonhumans: Moral status for them is not grounded

in moral personhood any more than it is grounded in metaphysical personhood.

MORAL STATUS IN THE ABSENCE OF PERSONHOOD

It is fortunate for animals and humans who lack moral personhood that moral

status does not require personhood of any type. The reason is that certain non-

cognitive andnonmoral properties are sufficient to confer ameasure ofmoral status.

At least two kinds of properties qualify a creature: properties of having the

capacity for pain and suffering and properties of emotional deprivation. As Jeremy

Bentham pointed out, the capacity to feel pain and undergo suffering is more

relevant to moral status for nonhuman animals than are cognitive properties.13 The

emotional lives of animals, though seldom discussed until very recently, are no less


important. Animals experience love, joy, anger, fear, shame, loneliness, and a broad

range of emotions that can be radically altered, distorted, or numbed by their

circumstances (Griffin, 1976; Masson & McCarthy, 1995; Orlans et al., 1998).

Nonpersons have many interests in avoidance of pain, suffering, and emotional

deprivation. In principle, the status of such an individual could be so morally

considerable as to outweigh certain moral rights and interests of persons. For

example, the interests of animals could override the inherently limited rights of

humans to do research, own zoos, run museums, and operate farms.

The injunction to avoid causing suffering, emotional deprivation, and many

other forms of harm is as well established as any principle of morality. This

injunction is fashioned to protect individuals because harm is bad in itself, not

because it is bad for members of a certain species or type of individual, and not

because the individual is or is not a moral person. Animals have interests in

avoiding harms other than those of pain, suffering, and emotional deprivation.

For example, they have interests in not being deprived of freedomofmovement and

in continued life. The range of their interests is beyond the scope ofmy arguments. I

have merely maintained that we have at least some obligations to animals indepen-

dent of their status as persons and that noncognitive, nonmoral properties that

confer moral status form the basis of these obligations. This conclusion holds

equally for humans lacking metaphysical and moral personhood.

WHICH ANIMALS HAVE RIGHTS?

Thus far I have not discussed whether moral status includes rights for animals other

than the human animal. In a well-known article, Carl Cohen maintains that a right

is a claim that one party may validly exercise against another and that claiming

occurs only within a community of moral agents who can make claims against one

another and are authorized to do so. These rights “are necessarily human; their

possessors are persons” with the ability to make moral judgments and exercise

moral claims. Animals cannot have rights, he says, because they lack these abilities

(Cohen, 1986, p. 865; 1990), which is to say that they lack moral personhood. Kant

seems to hold this view, which is also embraced by many contemporary contrac-

tarians. They regard us as having only indirect obligations to animals.

Though widely embraced, this view endangers animals and humans alike.

A better account is that both humans and animals can be rights holders regardless

of whether they are metaphysical or moral persons. This conclusion follows from

my arguments about the diverse bases of moral obligations, but those arguments

need to be combined with the widely accepted doctrine in law and morals that

rights are correlative to obligations. On this account, obligations always imply

corresponding rights if they are bona fide moral obligations, by contrast to merely


self-assumed obligations or personal moral ideals, such as “obligations” of charit-

able giving.14 “X has a right to do or to have Y” therefore means that the moral

system of rules (or the legal system, if appropriate) imposes an obligation on

someone to act or to refrain from acting so that X is enabled to do or have Y. The

language of rights is always translatable in this way into the language of obliga-

tions, and vice versa. For example, if a research investigator has obligations to

animal subjects to feed them and abstain from extremely painful procedures

during the conduct of research, then animal subjects have a right to be fed and

not to have the pain inflicted. More generally, if we have direct rather than merely

indirect obligations to animals, then animals have correlative rights. From this

perspective, a polarization into two fundamentally different types of theory,

premised on a difference about rights and obligations, makes no sense.

Correlativity requires that anyone who recognizes obligations logically must

recognize that animals have whatever moral rights correspond to those obliga-

tions. Since Cohen and most thoughtful persons believe in some range of human

obligations to nonhuman animals that derive from some source, it follows that

they accept the view that the animals have correlative rights.

Possession of a right is also independent of being in a position to assert the

right. A right-holder need not be the claimant in a particular case. For example,

small children and the mentally handicapped may not be able to understand or

claim their rights. Nonetheless, they possess them, and claims can be made for

them by appropriate representatives. Similarly, animals have all the rights correl-

ative to obligations that humans owe them, and they have these rights regardless

of whether they or any surrogate is in a position to exercise the rights.

Whatever the precise set of rights of animals and of humans who fall short of

moral personhood, and whatever their precise level of moral status, that set of

rights will not be the same as the set of rights enjoyed by moral persons. Because

bears and beagles lack the accountability and moral agency found in moral

persons, their rights are different. A theory of moral personhood should help us

understand why some entities have a full moral status, but the theory will not be

sufficiently powerful to exclude other entities from a partial moral status. This

point is not trivial, because some of the most important moral questions about

our uses of both humans and nonhumans—for example, as sources of organs and

as subjects of research—turn on the precise moral status of these individuals.

THE PROBLEM OF VAGUENESS IN THE CONCEPT OF PERSONS

Conceptual unclarity is another problem about theories of personhood that

deserves attention. Literature on the criteria of persons is mired in an intractable

dispute over a wide range of cases, including fetuses, newborns, the irreversibly


comatose, God, extraterrestrials, and the great apes. Facts about these beings are

not the source of the dispute. The problem is created by the vagueness and the

inherently contestable nature of the ordinary language concept of person,15 with

its commitments to a human individual composed of a rather open-textured set

of mental traits.

The vagueness of this concept is not likely to be dissipated by general theories of

personhood, which will invariably be revisionary of the concept. Theories typi-

cally reflect the concept’s vagueness and kindle more disagreement than enlight-

enment.16 They give us nomore than grounds for a claim that there are alternative

sets of sufficient conditions of personhood. The possibility of necessary and

sufficient conditions of person in a unified theory now seems dim. The concept

of person is simply not orderly, precise, or systematic in a way that supports one

general philosophical theory to the exclusion of another.

There is a solution to this problem of vagueness in the concept of person: Erase it

from normative analysis and replace it with more specific concepts and relevant

properties. I favor this option for bothmetaphysical personhood andmoral person-

hood because it would enable us to go directly to the heart of substantive moral

issues instead of using the oblique detour now made through theories of person-

hood. That is, we could inquire directly about the moral implications of possessing

specific nonmoral andmoral properties, such as reason andmoralmotivation, orwe

could discuss the substantive bases of ascriptions of rights. Questions about whether

fetuses can be aborted, whether xenotransplantation is permissible, and whether

anencephalics can be used in human experiments would then be recast in terms of

whether and, if so, on what moral grounds such actions can be performed.

My suggestion is not that we should abandon philosophical theories of meta-

physical persons and moral persons. My interest is exclusively in eliminating the

abuse of these theories in normative analysis, not in eliminating the theories

themselves.

CONCLUSION

I have said relatively little about specific normative problems or about the practical

implications of the conclusions I have reached, but not because these questions are

unimportant. I conclude with a comment on how very important they are.

Much has been made of the potential breakdown of the lines that have

traditionally distinguished human and nonhuman animals. If nonhumans turn

out to possess significantly more advanced capacities than customarily envi-

sioned, their moral status would be upgraded to a more human level. However,

this possibility remains speculative and may be less important than the thesis that

because many humans lack properties of personhood or are less than full persons,


they are thereby rendered equal or inferior in moral status to some nonhumans. If

this conclusion is defensible, we will need to rethink our traditional view that

these unlucky humans cannot be treated in the ways we treat relevantly similar

nonhumans. For example, they might be aggressively used as human research

subjects and sources of organs.

Perhaps we can find some justification of our traditional practices other than a

justification based on status as persons or nonpersons. However, if we cannot find

a compelling alternative justification, then either we should not be using animals

as we do or we should be using humans as we do not.17

References

Beauchamp, Tom L. 1992. “The Moral Status of Animals in Medical Research.” Journal of

Law, Medicine, and Ethics 20: 7–16.

Benn, Stanley. 1976. “Freedom, Autonomy and the Concept of a Person.” Proceedings of the

Aristotelian Society 76: 123–130.

_________. 1988. A Theory of Freedom. Cambridge: Cambridge University Press.

Bentham, Jeremy. 1948. The Principles of Morals and Legislation. New York: Hafner.

Biernbacher, Dieter. 1996. The Great Apes—Why They Have a Right to Life. Etica &

Animali 8: 143.

Buchanan, Allen, and Brock, Dan. 1989.Deciding for Others: The Ethics of Surrogate Decision

Making. Cambridge: Cambridge University Press.

Cohen, Carl. 1986. “The Case for the Use of Animals in Research.” New England Journal of

Medicine 315: 865–870, esp. 865.

_________. 1990. “Animal Experimentation Defended.” In The Importance of Animal

Experimentation for Safety and Biomedical Research, ed. S. Garattini and D. W. van

Bekkum, 7–16. Boston: Kluwer Academic.

Darwin, Charles. 1981. The Descent of Man. Princeton: Princeton University Press.

DeGrazia, David. 1996. Taking Animals Seriously: Mental Life and Moral Status. New York:

Cambridge University Press.

_________. 1997. “Great Apes, Dolphins, and the Concept of Personhood.” Southern

Journal of Philosophy 35: 301–320.

Dennett, Daniel. 1976. “Conditions of Personhood.” In The Identities of Persons, ed. Amelie

O. Rorty, 175–196. Berkeley: University of California Press.

Dworkin, Gerald. 1988. The Theory and Practice of Autonomy. New York: Cambridge

University Press.

Ekstrom, Laura W. 1993. “A Coherence Theory of Autonomy.” Philosophy and

Phenomenological Research 53: 599–616.

Engelhardt, H. Tristram, Jr. 1996. The Foundations of Bioethics. 2nd ed. New York: Oxford

University Press.

English, Jane 1975. “Abortion and the Concept of a Person.”Canadian Journal of Philosophy

5: 233–243.


Feinberg, Joel. 1989. “The Rights of Animals and Future Generations.” InAnimal Rights and

Human Obligations, 2nd ed., ed. Tom Regan and Peter Singer, 190–196. Englewood

Cliffs, NJ: Prentice Hall.

_________and Levenbook, Barbara Baum. 1993. Abortion. In Matters of Life and Death:

New Introductory Essays inMoral Philosophy, 3rd ed., ed. TomRegan, 195–234. New York:

Random House.

Francione,Gary L. 1995.Animals, Property, and the Law. Philadelphia: TempleUniversity Press.

Frankfurt, Harry G. 1971. “Freedom of the Will and the Concept of a Person.” Journal of

Philosophy 68: 5–20.

Frey, Raymond. 1988. “Moral Status, the Value of Lives, and Speciesism.” Between the

Species 4 (3): 191–201.

_________. 1996. “Medicine, Animal Experimentation, and the Moral Problem of

Unfortunate Humans.” Social Philosophy and Policy 13: 181–211.

Gallup, Gordon G. 1977. “Self-Recognition in Primates.” American Psychologist 32: 329–338.

Gervais, Karen G. 1986. Redefining Death. New Haven: Yale University Press.

Goodman, Michael. 1988. What Is a Person? Clifton, NJ: Humana Press.

Griffin, Donald R. 1976. The Question of Animal Awareness: Evolutionary Continuity of

Mental Experience (2nd ed., 1981). New York: Rockefeller University Press.

_________. 1992. Animal Minds. Chicago: University of Chicago Press.

Harris, John. 1985. The Value of Life. London: Routledge.

Locke, John. 1975. An Essay Concerning Human Understanding, ed. Peter Nidditch. Oxford:

Clarendon Press.

Lomasky, Loren. 1987. Persons, Rights, and the Moral Community. Oxford: Oxford

University Press.

Marras, Ausunio. 1993. “Pollock on How to Build a Person.” Dialogue 32: 595–605.

Masson, Jeffrey M., andMcCarthy, Susan. 1995.When Elephants Weep: The Emotional Lives

of Animals. New York: Delacorte Press.

Miles, H. L. 1993. “Language and the Orang-utan: The Old Person of the Forest.” In The

Great Ape Project, ed. Paola Cavalieri and Peter Singer, 52. New York: St. Martin’s Press.

Orlans, F. Barbara, et al. 1998. The Human Use of Animals: Case Studies in Ethical Choice.

New York: Oxford University Press.

Patterson, Francine, and Gordon, Wendy. 1993. “The Case for the Personhood of Gorillas.”

In The Great Ape Project, ed. Paola Cavalieri and Peter Singer, 58–77. New York: St.

Martin’s Press.

Pollock, John. 1989. How to Build a Person. Cambridge, MA: MIT Press.

Puccetti, Roland. 1969. Persons. New York: Harder and Harder.

Rachels, James. 1990. Created From Animals: The Moral Implications of Darwinism. New

York: Oxford University Press.

_________. 1993. “Why Darwinians Should Support Equal Treatment for Other Great

Apes.” In The Great Ape Project, ed. Paola Cavalieri and Peter Singer, 152–157. New York:

St. Martin’s Press.

Regan, Tom. 1983. The Case for Animal Rights. Berkeley: University of California Press.

Rodd, Rosemary. 1990. Biology, Ethics, and Animals. Oxford: Clarendon Press.


Sapontzis, S. F. 1987. Morals, Reason, and Animals. Philadelphia: Temple University Press.

Savulescu, J. 1994. “Rational Desires and the Limitation of Life-Sustaining Treatment.”

Bioethics 8: 191–222.

Strawson, Peter. 1959. Individuals. London: Methuen.

Tooley,Michael. 1972. “Abortion and Infanticide.” Philosophy and Public Affairs 2 (Fall): 37–65.

_________. 1983. Abortion and Infanticide. Oxford: Clarendon Press.

_________. 1984. “In Defense of Abortion and Infanticide.” In The Problem of Abortion,

2nd ed., ed. Joel Feinberg, 120–134. Belmont, CA: Wadsworth Publishing Company.

Warren,MaryAnne. 1973. “On theMoral and Legal Status of Abortion.”TheMonist 57: 43–61.

_________. 1991. “Abortion.” In A Companion to Ethics, ed. Peter Singer, 303–314.

Cambridge, MA: Blackwell Reference.

Notes

1. Other philosophers have used this or a similar distinction, but not as I analyze the

distinction (cf. Dennett, 1976, esp. 176–178; Feinberg & Levenbook 1993; Sapontzis 1987, 47ff).

2. A respectable case can be made that Aristotle, Boethius, Descartes, Locke, Hume, and

Kant either presupposed or argued for this position. However, Kant’s apparent inclusion of

moral autonomy renders him a borderline case, and other qualifications would need to be

made for some of these figures.

3. Tooley’s clarification of the distinction between the descriptive and the normative

functionsof “person” isuseful. For a concise andpersuasive account of the descriptive (factual)

andnormative (implying rights andduties) uses of the concept of person and the philosophical

importance of the distinction, see Biernbacher (1996, 143). However, neither account captures

the notion of moral personhood, which is more descriptive than normative. The normative

dimension is best understood in terms of the moral status of persons, irrespective of whether

that status is attributed on the basis of metaphysical or moral personhood. This point seems

generally overlooked in the relevant literature (see, e.g., Gervais, 1986, 181).

4. Throughout the histories of philosophy and law, there has been little resistance to the

postulate that animals have no moral or legal standing because they lack the properties that

confer moral status. Animals have been given almost no legal standing in British and

American systems of law, but questions of their moral status are far from decided (see

Beauchamp, 1992; DeGrazia, 1997; Francione, 1995, Chapter 4; Frey, 1988, esp. 196–197;

Rachels, 1990; Regan, 1983).

5. I do not deny the possibility of a theory of metaphysical personhood. My objections do

not apply to some of the early and influential metaphysical accounts in contemporary philo-

sophy, such as Strawson (1959) and Puccetti (1969). Locke, as cited earlier, is another example.

6. On the relevance and plausibility of robots and physical-mental systems that imitate

human traits, see Pollock (1989) and Marras (1993).

7. See Tooley (1972, 1983); also see Engelhardt (1996, Chapters 4, 6); Lomasky (1987);

Tooley (1984); and Warren (1973, esp. sec. 2; 1991, esp. 310–313). See, too, the articles and

bibliography in Goodman (1988).


8. There are more demanding theories than these second-order reflection theories.

Some theories require extremely rigorous standards in order to be autonomous or to be

persons. For example, they demand that the autonomous individual be authentic, con-

sistent, independent, in command, resistant to control by authorities, and the original

source of values, beliefs, rational desires, and life plans (see Benn, 1976; 1988, 3–6, 155f, 175–

183; see also Savulescu, 1994).

9. See Donald R. Griffin, Animal Minds (Chicago: University of Chicago Press, 1992).

10. Some measure of personhood is gained or lost over time as critical capacities are

gained or lost. These capacities are often gained or lost gradually, feeding the hypothesis of

degrees of personhood.

11. For the kinds of capacities and action that appear to constitute an exception, see the

study of gorillas in Patterson and Gordon (1993, esp. 70–71).

12. Darwin argues that moral sensitivity is itself the product of evolution. He maintains

that some humans display a high level and other humans a low level of moral responsive-

ness; the highest level of morality is reached when persons extend their sympathies beyond

their own group and indeed beyond their own species to all sentient creatures.

13. Bentham (1948, Chapter 17, sec. 1) reasons as follows: “If the being eaten were all,

there is very good reason why we should be suffered to eat such of them as we like to eat: we

are the better for it, and they are never the worse. They have none of those long-protracted

anticipations of future miserys which we have. The death they suffer in our hands

commonly is, and always may be, a speedier, and by that means a less painful one, than

that which would await them in the inevitable course of nature. . . . But is there any reason

why we should be suffered to torment them? Not any that I can see. Are there any why we

should not be suffered to torment them? Yes, several. . . .”

Donald Griffin has argued that no good reason exists to place a special weight on the

distinction between perceptual awareness in animals and a reflective consciousness. Griffin

(1992, esp. 248) proposes multiple levels of mentation shared across species, from basic pain

receptors to intentionality (see also DeGrazia, 1996; Rodd, 1990).

14. See Joel Feinberg’s (1989) argument that animals can have rights because they have,

or at least can have, interests that we are obligated to protect. For example, animals have the

right to be treated humanely, which follows from an obligation of justice. Feinberg and

others have argued that we have a strong moral obligation not to cause unnecessary

suffering in animals, and that correlative to this obligation are rights for animals—even

if the fulfillment of that obligation has the consequence of losing benefits for humans.

15. For an early and influential analysis of this problem, see English (1975); see also

DeGrazia (1996, esp. 305–315).

16. Some theories appeal to human characteristics, others to beings with moral status,

and still others to those with properties such as second-order volition (Frankfurt), moral

volition (Kant), and the capacity to experience pain and suffering (Bentham). Each account

is tied to some larger philosophical doctrine. Without judging the merits of the latter, it is

difficult if not impossible to judge the former.

17. My conclusion here has been influenced by private discussions with Raymond Frey

(see Frey, 1996); see also the engaging essay by Rachels (1993).


15

LOOKING BACK AND JUDGING OUR

PREDECESSORS

It seems widely believed, as the chairman of the Chemical Manufacturers

Association recently opined, that “You cannot judge people or a company based

on today’s standards or knowledge for actions taken 40 to 60 years ago” (“Ex-

owner,” 1994). The claim is that people cannot be judged or held responsible for

forms of waste disposal, research practices, types of marketing, and the like that

were common and rarely challenged a half-century ago. The Advisory Committee

on Human Radiation Experiments (ACHRE) has recently shown that this thesis

cannot be sustained, at least not without heavy qualification.

The Advisory Committee’s framework of distinctions and principles is found in

Chapter 4 of its Final Report, entitled “Ethics Standards in Retrospect” (ACHRE,

1995). I accept the major components of the moral framework of principles

presented by the Advisory Committee, but I will argue that the Advisory

Committee does not altogether adhere to the language and commitments of

that framework of principles in later chapters of the Final Report. My concern is

with how the Advisory Committee’s principled framework was used and how it

could and should have been used to make retrospective judgments about wrong-

doing and culpability. I will here limit my comments to plutonium cases (see

ACHRE, 1995, Chapter 5), and I will emphasize the University of California at San

Francisco (UCSF) case. However, these conclusions are generalizable to the entire

report, especially to the “Findings” in Chapter 17.

Although there are imperfections in the Advisory Committee’s use of its own

framework in reaching judgments of culpability, these shortcomings are sig-

nificantly offset by its persuasive and well-documented judgments of wrong-

doing. Furthermore, this committee’s weaknesses look like strengths when

compared to the more serious deficiencies in the Report of the UCSF Ad Hoc

Fact Finding Committee on World War II Human Radiation Experiments (UCSF,

1995). The latter report, I will argue, effectively has no framework and reaches no

significant judgments of either wrongdoing or culpability, although it was

positioned to do so.

STANDARDS FOR RETROSPECTIVE JUDGMENT

In its Final Report, the Advisory Committee presents a well-developed set of

distinctions, principles, explanations, and arguments pertinent to retrospective

moral judgments. It delineates “An Ethical Framework” (ACHRE, 1995 [GPO,

pp. 197–212; Oxford, pp. 114–124]), which I will call “the framework.” Its core, for

my purposes, is (1) a distinction between wrongdoing and culpability; (2) the

identification of three kinds of ethical standards relevant to the evaluation of

the human radiation experiments (basic principles, government policies, and

rules of professional ethics); and (3) an account of culpability and exculpation,

including mitigating conditions that are exculpatory. I will first review what the

Final Report says about these three parts of the framework, and then turn to

critical appraisal later.

The Wrongness of Actions and the Culpability of Agents

The distinction between the wrongness of actions and the culpability of agents

derives from the need to distinguish whether one is evaluating themoral quality—

in particular, the wrongness—of actions, practices, policies, and institutions or

evaluating the blameworthiness (culpability) of agents. The Advisory Committee

correctly concluded that past actions regarding the conduct of human radiation

research can be judged instances of wrongdoing without at the same time judging

the agents who performed them to be culpable. From the fact that an action was

wrong or that someone was wronged by the action, it does not follow that the

agent(s) who performed the actions can be fairly blamed, censured, or punished.

This distinction is of the highest significance and is appropriately given a promi-

nent position in the Advisory Committee’s Ethical Framework (ACHRE, 1995

[GPO, pp. 208–212; Oxford, pp. 121–124]).


Universal Moral Principles and Contemporaneous Policies and Rules

The Advisory Committee identifies six basic ethical principles as relevant to its

work: “One ought not to treat people as mere means to the ends of others”; “One

ought not to deceive others”; “One ought not to inflict harm or risk of harm”;

“One ought to promote welfare and prevent harm”; “One ought to treat people

fairly and with equal respect”; and “One ought to respect the self-determination

of others” (ACHRE, 1995 [GPO, p. 198; Oxford, p. 114]). These principles state

general obligations. The Advisory Committee held that they are widely accepted

and that the validity of the principles is not time-bound. A hundred years or a

thousand years ago would not alter their moral force (ACHRE, 1995 [GPO, p. 199;

Oxford, p. 115]). By contrast, policies of government agencies and rules of

professional ethics do not have this universal quality and are often specific as to

time and place (ACHRE, 1995 [GPO, pp. 214–221; Oxford, pp. 125–130]).

Culpability and Exculpation

Third, building on the wrongdoing–culpability distinction, the Advisory

Committee added an account of culpability and exculpatory conditions

(ACHRE, 1995 [GPO, p. 208ff; Oxford, p. 121ff]). The Advisory Committee

found that several factors limit our ability to make judgments about the blame-

worthiness of agents. These include lack of evidence; the presence of conflicting

obligations; factual ignorance; culturally induced moral ignorance, in which

“cultural factors . . . prevent individuals from discerning what they are morally

required to do” (ACHRE, 1995 [GPO, p. 209; Oxford, p. 121]); an evolution in the

delineation of moral principles; and indeterminacy in an organization’s division

of labor and designation of responsibility.

Although these conditions are exculpatory—that is, they mitigate or tend to

absolve of alleged fault or blame—satisfaction of the conditions does not always

exculpate and only two of the conditions affect judgments of wrongdoing (ACHRE,

1995 [GPO, p. 204; Oxford, p. 118]). The conditions are satisfied by degrees, and

exculpation can involve balancing several different considerations (ACHRE, 1995

[GPO, p. 210; Oxford, p. 122]). Presumably at the heart of the Advisory Committee’s

work is an examination of whether these exculpatory conditions were present in

particular cases in order to determine whether the persons involved are excul-

pated—that is, are free or relatively free of blame for what they did.

The Advisory Committee understood its first task to be that of evaluating the

rightness or wrongness of actions, practices, and policies. It emphasized the impor-

tance of discovering whether judgments ascribing blame to individuals or groups

can and should be made. It noted that “unless judgments of culpability are made

L O O K I N G B A C K A N D J U D G I N G O U R P R E D E C E S S O R S 263

about particular individuals, one importantmeans of deterring future wrongswill be

precluded” (ACHRE, 1995 [GPO, p. 212; Oxford, p. 123], emphasis added).

THE UCSF CASE

One case considered by the Advisory Committee, and the only one I will consider

in detail, originates at UCSF and eventuates in the Ad Hoc Committee report

previously mentioned. I will use this case as an example of the Advisory

Committee’s application of its framework to assess a case and as a means of

evaluating the UCSF Ad Hoc Committee’s judgments.1

Plutonium Injections from 1945 to 1947

The salient facts in this case are as follows. From April 1945 to July 1947, 17 patients

were injected with plutonium at three university hospitals—UCSF, the University

of Chicago, and the University of Rochester—and 1 patient was injected at Oak

Ridge Hospital in Tennessee. Three of the 18 (known as CAL-1, 2, and 3) were

injected at UCSF. Eventually allegations surfaced that the injections of plutonium

were toxic to the patients and that they never consented to involvement as subjects

of the research.

These patient-subjects were part of a secret research protocol initiated by the

Manhattan Engineer District, a government program responsible for the produc-

tion of atomic weapons. The purpose of this scientific research was to determine

the excretion rate of plutonium in humans so that the government could establish

safety levels and standards for workers who handled this radioactive element.

Although performed largely in secret, some of this work was known publicly as

early as 1951. It was discussed in the 1970s and 1980s, but did not generate a

substantial controversy until 1993, as a result of the work of a persistent investi-

gative reporter in Albuquerque, N.M.

The 1995 Report of the UCSF Ad Hoc Committee

On January 7, 1994, UCSF Chancellor Joseph Martin appointed an Ad Hoc

Committee to investigate these allegations by studying the history of UCSF

involvement. After a year of investigating documents and debating the issues,

this committee filed its report in February 1995. The Ad Hoc Committee con-

firmed UCSF involvement and confirmed that at least one of the three initial

patients had been included in the secret government protocol. The UCSF report


notes that the “injections of plutonium were not expected to be, nor were they,

therapeutic or of medical benefit to the patients” (UCSF, 1995, p. 27).2

The Ad Hoc Committee found that written consent was rare, disclosure

narrow, and the permission of patients not typically obtained even for nonther-

apeutic research during this period. They found that it is not known and could not

be known exactly what these research subjects were told or what they understood

(UCSF, 1995, pp. 26, 34). They also found that the word plutoniumwas classified at

the time; it is therefore certain that it was not used in any disclosures or explana-

tions that might have been made to patient-subjects. The committee noted that in

a recorded oral history in 1979, Kenneth Scott, one of the three original UCSF

investigators, said that he never told the first subject what had been injected in

him. Scott went on to say that the experiments were “incautious” and “morally

wrong” (UCSF, 1995, pp. 26–27).3

Nonetheless, the Ad Hoc Committee wrote, “At the time of the plutonium

experiments, [today’s issues about proof of consent] were not discussed. The

Committee is hesitant to apply current-day standards to another historical

period,” although “overall, the Committee believes that practices of consent of

the era were inadequate by today’s standards, and even by standards existing at the

time. . . . [T]he Committee believes that researchers should have discussed risks

with potential subjects” (UCSF, 1995, pp. 32, 34).

The AdHoc Committee concluded that this experimentation offered no benefit

to subjects but also caused no harm because patients did not develop any medical

complications. It also found that the experiments themselves were “consistent

with accepted medical research practices at the time” (UCSF, 1995, p. 33). The Ad

Hoc Committee concluded that the subjects were wronged and the experimenta-

tion was “unethical” if the patients did not understand or agree to the interven-

tions; in this event, they were wronged because their integrity and dignity were

violated. However, the Ad Hoc Committee found that since it could not establish

what the patients understood, no basis existed for saying that they actually were

wronged (UCSF, 1995, pp. 28–34).

In an appendix to the Ad Hoc Committee’s report, a lawyer and committee

member, Elizabeth Zitrin, concluded that even if the experiments were consistent

with accepted medical practices at the time, “it does not make them ethical. And

they were not consistent with the highest standards of the time articulated by the

government, the profession, or the public.” The chairman of the Ad Hoc

Committee, Roy Filly, responded that this comment by the lawyer held investi-

gators to an unrealistically high research standard.4 I will consider this issue

momentarily, but first one further development in this case merits attention.

One day after the Ad Hoc Committee filed its report in February 1995, it

received a recently declassified memorandum about this research that had been

written on December 30, 1946. The memo was written by the chief of the


Operations Branch, Research Division, at the Oak Ridge Atomic Facility. His

subject was the preparations being made “for injection in humans by doctors

[Robert] Stone [one of the three original UCSF investigators] and [Earl] Miller.”

The memo reports that:

These doctors state that the injections would probably be made without the

knowledge of the patient and that the physicians assumed full responsibility.

Such injections were not divergent from the normal experimental method in

the hospital and the patient signed no release (Chapman, 1946, p. 1).

This memorandum and other known facts in the UCSF case indicate, despite

an excessively restrained Ad Hoc Committee report (discussed later in this

essay), that these patients were seriously wronged, even if they were not physi-

cally or mentally harmed (ACHRE, 1995 [GPO, pp. 249–252, 256–258, 264–269;

Oxford, pp. 149–152, 154–155, 160–163]). The UCSF committee was unwilling to

reach conclusions about wrongdoing, preferring, like the chairman of the

Chemical Manufacturers Association, the position that it is too demanding to

judge people or institutions for actions taken a half-century ago. Nowhere did

the Ad Hoc Committee consider culpability, but its conservative conclusions

strongly hint at exculpation on grounds of the limited evidence available to the

committee.

No analysis of this case or the other plutonium cases will be adequate if it evades

examination of questions of wrongdoing and culpability. The Advisory

Committee’s framework makes this demand. But how does the Advisory

Committee fare in making the judgments its framework calls for, and does it

fare better than the Ad Hoc Committee?

THE ADVISORY COMMITTEE’S FRAMEWORK

A section entitled “Applying the Ethical Framework” in the Advisory

Committee’s Final Report evaluates “human radiation experiments conducted

between 1944 and 1974” (ACHRE, 1995 [GPO, pp. 212–221; Oxford, pp. 124–130]). I

will first examine what, in my judgment, the Advisory Committee could be

expected to have decided in light of its own framework and, second, ways in

which the Advisory Committee deviated from what could be expected.

The Advisory Committee’s framework and discussion of its application indi-

cate that it undertook three assignments:

1. to locate moral wrongdoing that violates either basic moral principles or

guidelines in the government policies and professional ethics of the period;


2. to place the violations in the context of mitigating or exculpatory condi-

tions; and

3. to decide which persons and institutions are culpable and which excul-

pated, as well as which lie on a continuum between the two.

Through a resourceful and historically innovative examination of the actual

policies of government agencies and rules of professional ethics of the period, the

Advisory Committee unhesitatingly addressed the first assignment:

[T]hese experiments were unethical. . . . [T]wo basic moral principles were

violated—that one ought not to use people as a mere means to the ends of

others and that one ought not to deceive others. . . . The egregiousness of the

disrespectful way in which the subjects of the injection experiments and their

families were treated is heightened by the fact that the subjects were

hospitalized patients . . . [leaving] them vulnerable to exploitation

(ACHRE, 1995 [GPO, pp. 267–268; Oxford, p. 162]).

This evaluation of wrongdoing contrasts with the vacillation and indecisiveness

in the UCSF Ad Hoc Committee report. Although, with the exception of the

memo, both committees had basically the same documentary evidence before

them, the ways in which that evidence was used for retrospective judgments is

striking. This contrast makes for an excellent bioethics case study of different

approaches to the evidence for retrospective moral judgment.

The Advisory Committee tackled the second assignment with the same decisive-

ness and conscientious use of historical documents. In the UCSF case and other

radiation cases, it noted that judgments of excusability, culpability, and the like

depend, at least to some extent, on whether proper moral standards for research

involving human subjects were acknowledged in government agency policies and in

the culture of medicine. If relevant standards and duties were entirely undeveloped

at the time, this lamentable circumstance becomes exculpatory for persons accused

of wrongdoing. Such circumstances would be very different from a situation in

which there existed well-developed and officially endorsed policies for human

subjects research.

The Advisory Committee determined, however, that exculpation does not

come easily, because policies that included vital elements that would today be

considered central to the ethics of research involving human subjects existed even

in the mid-1940s. For example, the Advisory Committee concluded from the

evidence that “it was common to obtain the voluntary consent of healthy subjects

who were to participate in biomedical experiments that offered no prospect of

medical benefit to them” and that ill subjects were not a relevantly different class

of subjects (ACHRE, 1995 [GPO, p. 217; Oxford, p. 127]). The Advisory Committee


also found that “even fifty years ago, [the six basic] principles were pervasive

features of moral life in the United States” that everymedical investigator could be

expected to observe (ACHRE, 1995 [GPO, p. 204; Oxford, p. 118]).

In light of these discoveries and conclusions, the Advisory Committee often

reacted to proposed exculpatory conditions with the same strength and decisive-

ness with which it reacted to violations of principles. An example is found in its

conclusion that the mitigating condition of culturally induced moral ignorance

does not apply to many government officials, because they simply failed to

implement or communicate requirements that were already established as their

responsibility: “The very fact that these requirements were articulated by the

agencies in which they worked is evidence that officials could not have been

morally ignorant of them” (ACHRE, 1995 [GPO, p. 215; Oxford, p. 126]).

Throughout its assessments under assignment 2, the Advisory Committee

relied on the aforementioned cluster of factors, including factual ignorance and

culturally induced moral ignorance, that limit the ability to make judgments of

agent culpability. These criteria and the conclusions drawn from them contrast

markedly with the work of the UCSF Ad Hoc Committee, which attempted to

place a maximal distance between the policies of the 1940s and those of the 1990s,

rather than to locate their similarities. The Ad Hoc Committee was willing to find

mitigating conditions that either exculpate or suggest exculpation at almost every

point in the trail of evidence, whereas the Advisory Committee placed the

evidence in a broader and more revealing context.

The mood and strategy shift, however, in the Advisory Committee’s handling

of the third assignment. There is a sharp deviation from what I hypothesize to be

the expected path the Advisory Committee would take. Instead of assessing

culpability and exculpation for individuals and institutions, the Advisory

Committee focused almost entirely on the wrongness of actions. Despite its

statement, previously quoted, that “judgments of culpability [should be] made

about particular individuals,” no such judgment is reached in the Final Report,

not about the plutonium cases or any other cases, apparently because the

Advisory Committee found the trail of warranting evidence to run out at this

point.

This outcome is surprising in light of the Advisory Committee’s assessment

that many of the individuals involved in these cases had an obligation to the

norms governing conduct and yet often failed to take them seriously. For

example, the Advisory Committee found that various physicians could and

should have seen that using sick patients as they did in the plutonium cases was

morally worse than using healthy people, for in so doing one was violating not

only the basic ethical principle not to use people as ameremeans but also the basic

ethical principle to treat people fairly and with equal respect (ACHRE, 1995 [GPO,

p. 219; Oxford, pp. 128–129]).


Following these indictments of the actions of physicians—and government

officials, in corresponding passages—the reader is poised for the next step: a

treatment of mitigating conditions that are exculpatory, followed by an assess-

ment of culpability or exculpation. Surprisingly, this analysis never develops, or at

least does not develop for the evaluation of specific individuals. Only in the final

sentence of Chapter 5 does the Advisory Committee hint at the issue, using a new

language of “accountability.” It determined that responsible officials at govern-

ment agencies and medical professionals responsible for the plutonium injections

were “accountable for the moral wrongs that were done” (ACHRE, 1995 [GPO,

p. 269; Oxford, p. 163], emphasis added).

Clearly these professionals were accountable (i.e., responsible and answerable),

but were they culpable (blameworthy and censurable)? The term accountablemay

also here be functioning to pronounce blame, but the terms do not have the same

meaning, and the subtlety of the point will escape even close readers. The Advisory

Committee makes no connection in its Final Report between being accountable

and being culpable; nor does it present an argument to indicate that in reaching

the aforementioned judgment of accountability the Advisory Committee is

blaming the individual physicians responsible for the injections or any of the

other parties mentioned.

Some of this lost ground is recovered in the Advisory Committee’s “Finding 11”

in Chapter 17, where it finally restores the language of blame and returns to its

earlier style of reaching decisive conclusions:

The Advisory Committee finds that government officials and investigators

are blameworthy for not having had policies and practices in place to protect

the rights and interests of human subjects who were used in research from

which the subjects could not possibly derive medical benefits

(nontherapeutic research in the strict sense). By contrast, to the extent that

there was reason to believe that research might provide a direct medical

benefit to subjects, government officials and biomedical professionals are less

blameworthy for not having had such protections and practices. We also find

that, to the extent that research was thought to pose significant risk,

government officials and investigators are more blameworthy for not

having had such protections and practices in place (ACHRE, 1995 [GPO,

pp. 787–788; Oxford, pp. 503–504]).

This passage is particularly important for understanding what the Advisory

Committee believed it could and could not conclude in light of the massive body

of evidence before it. The Advisory Committee apparently thought that it could

blame only individuals and groups of individuals in institutions such as govern-

ment agencies and the medical establishment. While this blaming is amorphous


and anonymous, the Advisory Committee managed to reach a critical general

assessment:

[G]overnment officials and biomedical professionals should have recognized

that when research offers no prospect of medical benefit, whether subjects are

healthy or sick, research should not proceed without the person’s consent. It

should have been recognized that despite the significant decision-making

authority ceded to the physician within the doctor-patient relationship, this

authority did not extend to procedures conducted solely to advance science

without a prospect of offsetting benefit to the person (ACHRE, 1995 [GPO,

p. 788; Oxford, p. 504]).

These forms of blame are improvements over the weaker conclusions in

Chapter 5 about the plutonium cases. They also, once again, contrast noticeably

with the inconclusiveness in the UCSF Ad Hoc Committee report, which

assesses no form of blame, general or particular. Nonetheless, nowhere in the

Advisory Committee’s Final Report is a named agent, other than the federal

government and the medical profession, ever found culpable. This is true not

only in the Advisory Committee’s discussion of the plutonium cases, but also

throughout the Advisory Committee’s report. This outcome is not what one

would have expected after the Advisory Committee’s indictments of the inten-

tional actions of wrongdoing mentioned throughout its Final Report, and it

raises questions of what might have been decided within the Advisory

Committee’s framework had it not held out for exceptionally high standards

of historical and testimonial evidence before assessing individual blame (as is

discussed in the Conclusion of this essay).

WHAT FINDINGS SHOULD HAVE BEEN REACHED ABOUT THE

VIOLATIONS OF STANDARDS?

I will now consider the judgments the Advisory Committee should have reached,

given its framework and objectives. In some cases, it did reach these judgments; in

others, it did not. Again the UCSF case will serve as the principal example. In

asking whether the UCSF investigators violated moral standards, we could be

asking one or both of two questions, as the Advisory Committee notes. A first

question is whether these investigators violated well-articulated rules of profes-

sional ethics or government policies. A second question is whether universal rights

or principles were violated. The Advisory Committee offers convincing answers to

both questions. However, its answer to the second needs assessment before

proceeding to the more important questions of mitigating conditions.


Were Universal Standards Violated?

The Advisory Committee notes that the first two of its principles were violated,

namely, that personsmay not be used asmeremeans to the ends of others and that

persons may not be deceived by others; elsewhere in the Report it adds a violation

of its fifth principle, “One ought to treat people fairly and with equal respect”

(ACHRE, 1995 [GPO, pp. 219, 267; Oxford, pp. 128–129, 162]). The claim that three

principles were violated is justified, but the mention of only three principles is

puzzling. The evidence assembled in the UCSF case, and in other cases before the

Advisory Committee, is sufficient to conclude that all six of the principles

identified by the Advisory Committee were violated.

The third principle is, “One ought not to inflict harm or risk of harm.” Given

what was known and not known about plutonium at the time, it is reasonable to

infer that the physicians involved placed their patients at risk of harm, though

perhaps risk at an uncertain level. Whether this principle was violated was a

matter of debate during the Advisory Committee’s deliberations, but the available

evidence seems to warrant a conclusion that violations of the principle did occur.

It is not credible that a physician could inject a patient with plutonium without

any awareness that doing so involved risk for the patient. The fourth principle is,

“One ought to promote welfare and prevent harm.” In the context of the

plutonium experiments, it does not appear that a serious attempt was made to

prevent harm from occurring and certainly there was no affirmative promotion of

welfare. There was an attempt to refine the methods used so as to reduce the

likelihood and magnitude of harm, but no more. Finally, the sixth principle is,

“One ought to respect the self-determination of others.” This principle was

violated beyond a reasonable doubt, because the investigators routinely ignored

the right to consent to bodily invasions following adequate disclosures.

The Advisory Committee thus implicitly recognized that all six of its principles

were violated in the plutonium cases, but its failure to reach the explicit conclusion

that all six principles were violated is not insignificant. The Advisory Committee

identified its set of principles in order to provide a comprehensive account of

what could be reasonably expected of persons. It aimed for a complete assessment

of past practices in order to reduce the risk of errors and abuses in future human

experimentation. The Advisory Committee insisted that, in light of these goals, a

“complete and accurate diagnosis requires not only stating what wrongs were

done, but also explaining who was responsible for the wrongs occurring”

(ACHRE, 1995 [GPO, p. 212; Oxford, p. 123]).

Underassessment of the number of principles violated and those responsible for

violations outcomes are an unfortunate outcome. Perhaps it is but a minor error

in the Advisory Committee’s report, but the same type of error afflicts the report

of the UCSF Ad Hoc Committee.5 If the Advisory Committee hoped to affect the


thinking of future committee members on committees such as the one at UCSF, it

would have been instructive to show that and how all six principles had been

violated.

What Findings Should Have Been Reached About Exculpatory Conditions?

Is the ineffectual professional ethics during the period in question a condition that

mitigates blame—an exculpatory condition? Yes. Does this condition erase the

wrongs done to research subjects? No. Judgments of wrongdoing are not affected

by exculpatory conditions. Does the ineffectual professional ethics exculpate the

agents involved? No. A weak professional ethics merely mitigates (i.e., tempers or

lessens the severity of) blame; it does not clear of blame or exculpate. I will address

exculpation in a moment. The immediate question is about exculpatory condi-

tions and how they are related to the void in professional standards: What

conditions are exculpatory? Why are they exculpatory? To what degree do they

exculpate?

As previously noted, the Advisory Committee identified a set of exculpatory or

mitigating conditions: lack of evidence; the presence of conflicting obligations,

including obligations to protect national security; factual ignorance; culturally

induced moral ignorance; an evolution in the delineation of moral principles; and

indeterminacy in an organization’s division of labor and designation of respon-

sibility. To these conditions we might add that blame could be mitigated by

culturally induced misunderstandings, by a person’s good character, and by

what in law is called “excusable neglect,” which is neglect caused by an unavoid-

able hindrance or accident. Only three of these conditions need assessment here

because of the role they play in the plutonium cases: (1) factual ignorance, (2)

culturally induced moral ignorance, and (3) obligations to protect national

security.

Factual Ignorance

The Advisory Committee never argues that a significant measure of nonculpable

factual ignorance was operative in the UCSF case. The Advisory Committee holds

that claims of nonculpable factual ignorance can too easily function as an excuse

for wrongful actions and, by inference, as an excuse for failures to make retro-

spective moral judgments:

[J]ust because an agent’s ignorance of morally relevant information leads

him or her to commit a morally wrong act, it does not follow that the person


is not blameworthy for that act. The agent is blameworthy if a reasonably

prudent person in that agent’s position should have been aware that some

information was required prior to action, and the information could have

been obtained without undue effort or cost on his or her part (ACHRE, 1995

[GPO, p. 208; Oxford, p. 121]).

Culturally Induced Moral Ignorance

A claim of culturally induced moral ignorance lacks credible backing in the UCSF

case. Even if we grant that there were no strong cultural incentives to abstain from

the research and little instruction in medical ethics, it is not too much to expect

these physicians to have been aware that their actions required a justification

other than the experiment’s utility for others. Never in the history of civil

medicine has it been permissible to exploit patients by using them to the ends

of science in nontherapeutic research that carries risk of harm. In the plutonium

cases, the problem is not merely that no informed consent was obtained. No

consent at all was obtained.

Culturally induced beliefs, such as the belief that consent is not morally

required in a hospital setting, are most likely to constitute a valid excuse for

wrong actions when alternative views are unavailable or are not taken seriously in

the context. But alternative views were available and were considered matters of

the utmost significance in sources available to the relevant parties. It was known

or easily knowable at the time that (1) a debate had occurred during the mid-1940s

about experimentation in Nazi Germany; (2) the American Medical Association’s

Judicial Council had sided in 1946with what would soon be the Nuremberg thesis

that voluntary consent to participation in research is essential; (3) the Hippocratic

tradition required physicians to put the care of patients first, not to deviate

radically from accepted therapies, and not to risk harm to patients through

nontherapeutic interventions; and (4) there was a long tradition of post-

Hippocratic writings in medical ethics that included Thomas Percival, Claude

Bernard, and Walter Reed, each of whom recognized nontherapeutic experimen-

tation as valid only if subjects had consented. Thus, requirements such as volun-

tary consent to experimentation and protection against harmful interventions

had long been present in the medical community and even were present in some

government policies traceable to the early 1940s (see Lederer & Moreno, 1996).

In light of this history, the UCSF physicians could not plausibly appeal to

nonculpable moral blindness, because they and the officials at their institutions, as

well as responsible higher officials in medicine, could have been expected to

remedy contextual moral ignorance. There was ample opportunity for


remediation of inadequate moral beliefs and therefore culpability for the con-

tinuance of those beliefs. The excuse of nonculpable ignorance, then, is not

credible.

Conflicting Obligations—The National Security Exception

The matter of mitigating conditions is more complicated than nonculpable

ignorance, because it might be argued that there was reason to believe that the

research constituted a justifiable exception to ordinary physician obligations and

government policies requiring compliance with established standards. In the

UCSF case and others considered by the Advisory Committee, obligations to

protect national security might be viewed as conflicting with and overriding

obligations to protect human subjects. The so-called “national security excep-

tion” suggests that in order to survive as a nation and preserve a culture of

freedom, we can justifiably forfeit some measure of individual rights and inter-

ests—a classical utilitarian justification that promotes the public interest by

asking for some sacrifice on the part of individual citizens. The Cold War

struggle in the late 1940s could magnify the importance of this proposed excep-

tion; perhaps it could even serve as justifying the research done with human

subjects.

The Advisory Committee considers this argument and rightly blunts its force.

The Advisory Committee maintains that appeals to national security would have

unjustifiably caused investigators to lose sight of firm requirements of voluntary

consent. Those requirements could have been satisfied by asking subjects for their

permission after telling them that they would be injected with a radioactive

substance that might be dangerous and would not be beneficial, but would help

protect the health of persons involved in the war effort.

The culpability of agents in the plutonium cases might be mitigated by their

conscientious and understandable interpretation of the need to protect workers in

projects of great national significance, leading them to authorize or to perform the

research. Government officials and possibly the physicians with whom they

contracted could perhaps be found blameless because of the massive confusion

surrounding the ColdWar commitments and their general lack of familiarity with

research medicine, but the Advisory Committee rightly rejected the plausibility of

this claim, especially for government employees.

In some respects this defense is even less plausible for physicians. The special

nature of the patient–physician relationship places a more stringent obligation on

physicians to attend to the welfare of the patient, and not merely their patients,

but any patient with whom they are professionally involved. In clinical circum-

stances, patients defer to their physicians’ judgment, and it seems transparent that


these physicians capitalized on this deference and failed to adequately protect the

welfare of their patients, a violation of the Advisory Committee’s fourth principle.

The Advisory Committee also discredited the thesis that national security was

ever formally invoked to justify these research efforts.

[I]n none of the memorandums or transcripts of various agencies did we

encounter a formal national security exception to conditions under which

human subjects may be used. In none of these materials does any official,

military or civilian, argue for the position that individual rights may be

justifiably overridden owing to the needs of the nation in the Cold War

(ACHRE, 1995 [GPO, p. 206; Oxford, p. 120]).

In short, the evidence does not indicate that the agents themselves viewed their

actions in this light, and even if they did, there were alternatives to the forms of

exploitation of patients that occurred in the plutonium cases.

Culpability or Exculpation?

The final problem is whether judgments of exculpation or judgments of culp-

ability follow from the foregoing analysis. Were the mitigating conditions suffi-

cient to exculpate the agents? To this question, the answer is, emphatically, “No.”

Weak training in professional ethics, embryonic federal policies, and parallel

forms of ignorance, together with other mitigating conditions, temper the reach

and level of possible judgments of blame, but they are not sufficient conditions of

exculpation. These conditions are only weakly exculpatory. Violation of the six

basic universal principles is, by itself, sufficient for a judgment of culpability.

Any intentional act of taking patients who were seriously ill and placing them at

risk without their knowledge or consent in nontherapeutic research indicates

culpability. The evidence suggests both that the physician-investigators know-

ingly exploited these patients and that sufficient opportunities existed for physi-

cians to obtain relevant information in order to determine whether their actions

were warranted. Even if responsibilities were not clearly assigned to individuals in

institutions, and even if the effort involved numerous persons, sufficient guide-

lines and historical precedents still existed to make judgments of culpability. That

federal officials and physicians associated with the plutonium cases were culpable

seems to follow from the Advisory Committee’s thesis that if the means to over-

come cultural biases or relevant forms of ignorance are available to an agent and

the agent fails to take advantage of these means, then the agent is culpable.

In reaching such conclusions, the Advisory Committee was right to insist that

we can and should assess persons other than the UCSF investigators, including


persons in positions of authority or responsibility for initiating the research and

for overseeing it—government authorities as well as those with oversight respon-

sibilities in medical and research institutions. Those responsible for setting,

implementing, and overseeing standards for the conduct of research are at least

as culpable as those who conducted the research.

CONCLUSIONS

All things considered, the performance of the physicians and government officials

involved in the plutonium cases is inexcusable. However, the fairest conclusion in

these cases may be that the various exculpatory conditions excuse the agents

involved to some degree. The moral culpability of agents admits of degrees, and

there are many degrees on the scale, depending on what is known, what is

believed, what is intended, and what is widely recognized and disseminated. For

example, there are degrees of culpable moral ignorance in these cases, and it

would be difficult to pinpoint the degree of culpable ignorance for any particular

agent.

Perhaps it is enough to conclude that whatever the degrees of culpability,

culpability there must be. Amilder conclusion is that it is enough to judge persons

or institutions deficient in conduct, which carries a loss of status and reputation,

without attaching the stigma of blame or inflicting punishments such as formal

censure, invalidation of a license, or fines. Which among these possible conclu-

sions did the Advisory Committee reach, and was it correct to take the course it

did?

The Advisory Committee’s even-handed Findings 10 and 11 (ACHRE, 1995

[GPO, pp. 785–789; Oxford, pp. 502–504]) provide a clue. Here the Advisory

Committee determines that physicians and government officials were “morally

responsible in cases in which they did not take effective measures to implement”

available government and professional standards. The most favorable interpreta-

tion of these findings, in light of the rest of the Final Report, is that the Advisory

Committee is blaming government officials and associated physicians for moral

failures. Despite its relatively polite language of “accountability” and “responsi-

bility,” the Advisory Committee’s view strongly suggests that government officials

and physicians in positions of leadership were culpable for serious moral failures.

In the attempt to be even-handed and not to stretch beyond the evidence—a

key consideration in the Advisory Committee’s deliberations and findings

(Buchanan, 1996)—the Advisory Committee reached conclusions that do not

extend all the way to the culpability of individual physicians and government

officials. In eschewing questions of individual culpability, the Advisory

Committee risks reproach for indecisiveness, much as I have criticized the


UCSF Ad Hoc Committee, and some reproach is in order. However, in assessing

the fairness of such criticism of both the Advisory Committee and the UCSF

committee, these committees did have a responsibility not to overstep the evi-

dence, a task that becomes increasingly difficult in assessing the culpability of

particular individuals.

It would also be rash to judge the Advisory Committee harshly for not inves-

tigating each government official and physician who might be blamed for wrong-

doing. The Advisory Committee existed for little more than a year and did not

have the means for a broad-scale investigation of individuals. Nonetheless, I

believe that sufficient evidence exists and was available to the Advisory

Committee for judgments of culpability in the case of a number of individuals

involved in the human radiation experiments. In the UCSF case alone, investi-

gator Kenneth Scott himself testified to physician culpability during the afore-

mentioned interview. The evidence also indicates that principal investigator

Robert Stone was guilty of the kind of negligence, errors of judgment, and

moral failures that are sufficient for culpability.

If we cannot judge particular persons blameworthy in a case as clear as that of

the UCSF plutonium injections, it is hard to see how to stop short of either

exculpation or paralysis of judgment in a great many cases of serious past moral

wrongs. This problem of line drawing cannot be dismissed, as the Advisory

Committee recognized, on grounds that it is too difficult for human judgment.

We should be cautious but not incapacitated in the work of retrospective moral

judgment. Cautiousness in making claims to have sufficient evidence of culp-

ability is a virtue, but suspending judgment in the face of sufficient evidence is

simply a failure to make the proper judgment.

The matter is complicated by questions of the proper criteria of sufficient

evidence. More than one standard of evidence can be defended, and significant

epistemological problems surround the defense of one standard over another. All

evidence gathering assumes a theory of what counts as evidence, and two or more

theories may support competing standards, making assertions of sufficiency

inherently contestable. While this problem is of indisputable relevance to assess-

ments of culpability, it does not provide an adequate reason to doubt the avail-

ability of a reasonable standard for making such assessments. In the present case, I

do not believe that the standard would vary significantly from those used by the

Advisory Committee for assessing wrongdoing.

The bigger problem is the amount of time it would have required to assemble

the evidence properly in each particular case, given the lapse of 50 years and the

many gaps of information as to what did and did not happen. Time was not

available for all cases to be handled appropriately, nor could all the desired

evidence have been acquired (Buchanan, 1996). Since it generally is less time

consuming to assemble evidence of sufficient quality to assess culpable


nonperformance within the professions than to assemble comparable evidence for

culpable wrongdoing on the part of each individual, it is easy to understand why

the Advisory Committee reached the conclusions it did. But with more careful

collection and sifting of the evidence in individual cases, the quality of the

evidence may turn out to be as good as the evidence of general institutional

culpability in medicine and government.

A judgmental person is a fool, but fear of rendering a foolish judgment some-

times induces an unwarranted reserve. Such restraint appears to have unjustifi-

ably inhibited decision making by the UCSF Ad Hoc Committee. The Advisory

Committee cannot be similarly evaluated, because it was decisive beyond what

might reasonably have been anticipated in light of its broad and diverse mission.

However, it seems likely that the Advisory Committee’s revealing findings are no

more than a starting point for judgments of individual culpability that we can

expect and should encourage in the ongoing work on the subject of the human

radiation experiments.

References

Advisory Committee on Human Radiation Experiments (ACHRE). 1995. Final Report.

Washington, DC: U.S. Government Printing Office. Subsequently published as The

Human Radiation Experiments. New York: Oxford University Press, 1996. Pagination

for both volumes is provided in the textual citations.

Buchanan, Allen. 1996. “The Controversy over Retrospective Moral Judgment,” Kennedy

Institute of Ethics Journal 6 (1996): 245–250.

Chapman, T. S. 1946. “Memorandum: To Area Engineer, Berkeley Area (30 December

1946).” ACHRE No. DOE-112194-D-3.

Davidson, Keay. 1995. “Questions Linger on 1940s UCSF Plutonium Shots,” The San

Francisco Examiner (23 February 1995): A6.

“Ex-owner ofToxic SiteWinsRulingonDamages.” 1994.NewYorkTimes (18March 1994): 5B.

Lederer, Susan E., and Moreno, Jonathan D. 1996. “Revising the History of Cold War

Research Ethics,” Kennedy Institute of Ethics Journal 6: 223–237.

UCSF. University of California at San Francisco. 1995. Report of the UCSF Ad Hoc Fact

Finding Committee on World War II Human Radiation Experiments (February, 1995,

unpublished but released to the public.)

Notes

1. In presenting the facts in the UCSF case, some parts of my discussion derive from the

Advisory Committee’s Final Report, but most derive from the AdHoc Committee’s Report.

All the facts were well known to the Advisory Committee, which discusses the UCSF case in

Chapter 5.


2. The amount of plutonium injected was approximately 0.1% of the LD50 in rats and

0.35% of the LD50 in dogs.

3. The oral history was conducted by medical historian Sally Hughes.

4. See Elizabeth A. Zitrin and Roy A. Filly, UCSF, “Report of the UCSF Ad Hoc Fact

Finding Committee,” Letters, Appendices. Filly’s response is reported by Keay Davidson

(1995).

5. For an example of this error, see the personal statement by UCSF committee member

Mack Roach III (UCSF 1995, Appendices).


I NDEX

Abandonment of patients, 40, 87Abuse (of patients and subjects), 74, 116,

133, 136, 271Accessto clinical trials, 36

to drugs, 26

to health care, 40, 133–37to medical records, 74

to results of research, 26

See also Allocation of scarce resources;Justice; Research

Accountability, 250, 255, 269, 276Ad Hoc Committee of UCSF (Plutonium

experiments), xxv, 261–78Addiction, 92, 105, 111Advance directives, 46, 177, 183Advisory Committee on Human Radiation

Experiments, xii, xxv, 29, 261–76Agency. See AutonomyAIDS, 26Allocation of scarce resources, 154,

212, 216. See also Access; JusticeAllowing to die. See Euthanasia; Killing and

letting die; Refusal of treatmentAmerican Medical Association (AMA), 14,

48, 50, 52, 58, 273Analogy, its place in methods of ethics,

160–64, 171. See also CasuistryAnimals (nonhuman), xxiv–xxv, 91, 93–96,

250–56, 259–60. See also Moral statusAppelbaum, Paul, 39

Applied ethics, xi, xiv, xxii, 47, 211–226, 243Aristotle, 95, 216, 259Arnold, Denis, xiArras, John, ix–x, xx

Autonomyautonomous actions contrasted withautonomous persons, 79, 89, 93, 96

autonomous authorization, 63–65,79–80

autonomy-limiting principles, 102–06,129

balanced with beneficence. SeePaternalism

the concept of, 79–80, 89, 239–41its connection to informed consent,63–65, 82, 88, 142–44

degrees of, 88, 252diminished, 24, 37, 70, 80, 137, 142–44Dworkin’s theory of, x, xxiv, 89, 219,

240–41, 250–51Frankfurt’s theory of, xviii, 89–92, 95, 99,240, 251

intentionality a condition of, 63–66,71–72, 80, 83–85, 88, 94–95, 270

metaphysical contrasted to moral, 79,94–96

paradigm cases of, 82

principle of respect for, xvii–xviii, 7,24, 36–38, 42, 46, 64–65, 74, 79–82,90–93, 97, 104, 109, 111–12, 166, 215,220, 240

theories of, 81–97, 240understanding a condition of, 37, 60,63–65, 68, 83–86, 88, 93, 251–52

voluntariness a condition of, 86–88, 90See also Advance directives; Informedconsent; Kantian theory; Liberty;Paternalism; Refusal of treatment;Suicide

Baier, Kurt, 196

Balancing moral normswith autonomy and beneficence, xviii,

27, 103–06, 110–12, 117with benefits, costs, risks, 22, 36, 39–40of justice, 140

processes or methods of, 28, 158–59,161–62

specification, relation to, 182–83in utilitarianism, 25–26, 110–11when different interests are involved, 27,

39, 75, 154–55, 263See also Paternalism; Specification

Battin, Margaret (Peg), xiBecker, Gerhold, xiiBelmont Report, Theits analytical framework of principles,

xv–xvi, 19, 21–24, 27, 29as applied research ethics, 9, 21authorized by U.S. Congress, 4, 13, 18beginnings of, 3–7, 9and casuistry, 10–12, 28connection to Principles of Biomedical

Ethics, ix–x, xiii–xv, 6–7historical origins of its principles, xiv–xv,

6, 9, 19, 36legacy and influence, 29–30scandals that motivated the work, 19–21,

56–57as universal research ethics, 23–24and utilitarian justifications of research,

25–26its weaknesses, deficiencies, and

unclarities, xiv–xv, 24–28See also National Commission for the

Protection of Human Subjects;Principlism and principles (inbioethics)

Beneficenceits balance with costs and risks, xviii, 22,

36, 39–40concept of, 24–25, 39–40and futility (nonbeneficial treatment),

122–24, 242. See also Killing andletting die

in the Hippocratic tradition, 35, 50–52,127, 156, 158, 231, 238, 273

ideals of, 40–41, 169. See also Ideals(moral)

moral principle(s) of, 4, 7–11, 21–22,24–25, 27, 39–40, 156, 163, 166, 215

nonmaleficence contrasted to, 38–40,169–70, 197

obligatory and ideal contrasted, 40–41,169–70

and paternalism, 52, 101, 103–06, 110–17and physician-hastened death, xi,120–29

See also Nonmaleficence; Paternalism;Principlism and principles (inbioethics); Utilitarianism

Benevolence, 39, 52, 103, 156, 166. See alsoBeneficence; Paternalism; Virtue(s)and virtue theory

Bentham, Jeremy, 253

Best interest(s)and paternalism, 103, 105, 112pursuit of patients’, 21, 74, 158See also Paternalism

Biascultural, 275

judgment affected by, and boundedrationality, 108, 195

measures protecting against, 156

patients’ (informational), 62

in science and method, xix, 132, 163, 179,203–04

See also Discrimination; PrejudiceBlame and blameworthiness, 96, 117, 122,

253, 262–63, 269–77; See alsoCulpability and wrongdoing

Bok, Sissela, x, 219Brain-damaged persons, 96, 253Breast implants (and the FDA), 115

Brock, Dan, xi, 29, 219Buchanan, Allen, x, xii, 219Bushkoff, Stanley, 55

Callahan, Daniel, x–xi, 221Campbell, Alasdair, xCapron, Alexander, 5, 70, 219Case-based accounts of ethics. See Case

methods in law, business, andphilosophy; Casuistry; Justification

282 I N D E X

Case methods in law, business, andphilosophy, xiii, 11, 212, 221–26

Casuistryanalogical reasoning in, 160

as an alternative paradigm toprinciplism, xx, xxiv, 10–12, 28, 153,160–64

problems with, 162–64, 233Roman Catholic, 177

Categorical imperative, 221. See also Kant;Kantian theory

Character. See Virtue(s) and virtue theoryCharity, 39, 88, 255. See also BeneficenceChildress, James F., ix–x, xiii–xvi, xx–xxi,

3–7, 10, 36, 40–44, 153–54, 157, 163–71,178, 184–85, 189ff

Christakis, Nicholas, 232–33Clinical trials, xix, 26, 55, 69–70, 132, 136,

138, 141. See also ResearchClouser, Danner, ix–x, xiii, xx, xxiii, 26–27,

167–71, 190, 218–19, 231, 233Coercion, 69–72, 80, 86–87, 102, 116, 137–39,

143, 219Coherence theoryas method of justification, xvi, 45–47,159, 183–85

and particular moralities, 182

reflective equilibrium a form of, 47, 159,182–83, 189, 193–94, 225, 230, 232–33,236–37

role in theories of autonomy, 80–81, 90Coma, 93, 128, 256Common moralitychange and stability in, 185–87, 198–200the definition of, 156–57, 175–76, 196–98empirical justification of claims about,xxii, 178–81, 183, 186, 191, 194, 200–04

Gert’s account of, xxi–xxii, 167–71,189–201

and morality in the descriptive sense,192, 194, 197–200

and morality in the normative sense, xi,xvi, xxi, 192–200, 203–04

normative justifications of, 43,180–82

objectives (goals) of the, xxi, 42–43, 169,171, 175–81, 185, 187, 192–93

and particular moralities, xxi, 6, 47, 175,177–78, 233

pragmatic justification of, 179, 192–94,199–200

and specification, 45–56, 157, 159,182–85

as the universal morality, xxi, 6,23–24, 28, 42–43, 161, 164, 171, 175ff,190, 196, 200, 202, 231–35, 263,271–72, 275

Compassion, 125, 165–65Competence and incompetenceincompetent persons, 9, 22, 24–25, 62,

73, 80in making decisions, 56, 58, 61–65, 72–75,135

and paternalism, 106–07, 113, 136in specifying autonomy, 46, 127, 182See also Autonomy; Moral status;Surrogate decision makers

Comprehension, 21, 61, 63, 215. See alsoInformed consent

Compromise, xx, 13, 23, 69, 71, 85, 106, 116,132–34, 142–45, 155, 158, 161, 184, 220

Conceptual analysisof fundamental moral norms, xxii, xxiv,194–98, 204

of killing and letting die, 122–25, 241–42Confidentiality, 44–45, 154, 156–57, 213–143Conflict of interest, xix, 82, 134, 158, 177Conscience, 217, 253Conscientiousness, 46, 167, 176Consensus, xviii, 28, 52, 55, 65, 101, 155,

180–81, 185, 231, 239Consent. See Informed consentConsidered judgments, 47, 97, 156, 159, 179,

193, 222, 225, 236. See also Reflectiveequilibrium

Constraining situations (or “coercivesituations”), xix, 71–72, 138–41, 149–50.See also Coercion; Manipulation

Contract(s), 72, 80, 139, 177, 181Convergence of theories, 230, 241Courage, 4, 7, 22, 59, 68, 120, 144, 161,

169, 197Craimer, Avi, ixCrouch, Robert, x

I N D E X 283

Culpability and wrongdoing, xiv, xxvi,261–63, 266–69, 274–78. See also Blameand blameworthiness

Cultural Relativity. See Pluralism andrelativism (moral)

Culturally induced moral ignorance, 263,268, 272–74

Culver, Charles, x, 15, 219, 233Customs, 67, 181, 185

Damages, 66, 68Daniels, Norman, xi, xxiii–xxiv, 189,

216–17, 230, 236–37, 240Darwin, Charles, 96, 253Davidson, Arnold, xiDavis, Wayne, ixDeception, 52–53, 192, 196, 214, 263, 267, 271.

See also LyingDeGrazia, David, ix, xi, xxi, 15–16, 49, 175,

180, 182–85, 188, 250Dementia, 96, 134, 247, 253Deontological theory, 27, 220. See also

Kant; Kantian theoryDepression, 80, 107Descriptive ethics, 180–81, 187, 192, 194,

197–200Dignity, xix, xxiv, 9, 22, 95, 100, 135, 265Dilemmas, xxiii, 12, 140–41, 172, 183, 223,

229

Disabled persons, 7, 69, 134, 204. See alsoVulnerabilities; Vulnerable personsand groups

Discernment (virtue of), 160, 166Disclosureits connection to law and liability,

66–68, 73and informed consent, 50–55, 58, 60–63,

142–44, 157–58, 219–20nondisclosure of pertinent information,

37, 65, 114–15, 239, 265, 271See also Confidentiality; Informed

consent; VeracityDiscretion, 19, 25, 44, 155Discrimination, 136, 156, 199, 214. See also

Bias; Disabled persons; Justice;Prejudice; Vulnerabilities; Vulnerablepersons and groups

Distress, 114, 128, 134, 136, 141–42, 235.See also Harms; Pain

Distributive justice, 41, 145, 217, 230, 236.See also Justice

Donagan, Alan, 6, 43, 190, 230Double-blind studies, 55. See also Clinical

trialsDworkin, Gerald, x, xxiv, 89, 219, 240–41,

250–51

Economically disadvantaged subjects, xi,xix–xx, 22, 132–46. See also Justice;Poverty; Vulnerabilities; Vulnerablepersons and groups

Egalitarian theories, 41

Emanuel, Ezekiel (Zeke), xEmergencies (and emergency medicine),

73, 112Emotions, 72–73, 86, 94, 96–97, 166,

253–54Empirical justification (in the study of

ethics), xxii, 47, 55–56, 59–60, 146,178–81, 183, 186, 191, 194, 200–04, 249

Engelhardt, H. Tristram, ixEqual consideration, 41, 185Equity, 144

Ethics of care, 172. See also Virtues andvirtue theory

Ethnicities, 186

Euthanasia, 46, 130, 213, 216, 241, 247.See also Killing and letting die;Physician-hastened death

Evil, 169–70, 192, 196–98, 216Exculpatory conditions, xxvi, 262–77ExperimentationAdvisoryCommittee onHumanRadiationExperiments, xii, xxv, 29, 262–76

on drugs, 55, 72,the Jewish Chronic Disease Hospitalcase, 56–57

Nazi, 20, 25, 56non-therapeutic, 37

involving prisoners, 72, 116–17Tuskegee syphilis case, 19–20, 25,

57, 71Walter Reed’s Yellow Fever, 53

See also Clinical trials; Research

284 I N D E X

Exploitation, xi, xix, 69–70, 116, 132–36,139–42, 145–46, 267, 275. See also Justice

Faden, Ruth, v, ix–xii, xvi, 42, 53, 56, 63,73–74

Feinberg, Joel, x–xi, 102, 106, 219Fetus(es)moral status of, 93, 96, 253, 255–56research involving, 19

Fidelity, 24, 105, 154, 167, 176Food and Drug Administration (FDA), 115Foot, Philippa, 196

Frankena, William, 6, 23Frankfurt, Harry, xviii, 89–92, 95, 99,

240, 251Freeman, Eugene, xiFraud, 57, 177Frey, R. G., xiiFutility (medical), 122–24, 242. See also

Killing and letting die

Gender roles, 178

Gert, Bernard, ix–xi, xiii, xx–xxii, 26–27,167–71, 189–205, 212, 214, 216, 218–20,230–31, 233, 237

Gewirth, Alan, 230

Gillon, Raanan, x, xiii, xviGlobalization and global ethics, 133, 230Goldman, Alvin, 83–84Gorovitz, Samuel, ix, 243Government. See RegulationsGrady, Christine, x, 29Great apes, 186, 250. See also Animals

(nonhuman)Green, Ronald, xi, 167Guardians, 73, 81, 104, 115

Handicapped persons, 105, 255. See alsoDisabled persons; Vulnerabilities;Vulnerable persons and groups

Harmsbalancing benefits and, 22, 35, 39, 127beneficence and preventing or removing,7, 40, 43, 109, 113, 127, 138, 176

the harm principle, 101–03, 106lessening, as the goal of morality, 192,196–98, 201

and negligence, 121. See also Negligencenonmaleficence and not causing, 11, 22,36, 38–39, 51, 73–74, 101, 127, 136, 154,156, 169, 192, 216, 253–54, 263, 266,271–73

paternalism and prevention or reductionof, 101–04, 106–07, 110–15

as setbacks to interests, 71–72, 87, 135, 139See also Nonmaleficence

Harris, John, x, xiiHegel, G. W. F., 225

Herissone-Kelly, Peter, ixHershey, Nathan, 55

Hepatitis, 20, 57Hippocratic tradition and Oath, 35, 50–52,

127, 156, 231, 238, 273Hobbes, Thomas and the Hobbesian

tradition, 178, 192–97Honesty, 4, 176Hooker, Worthington, 52–53Hope, fostering in health care, 52, 70–71, 165Hospers, John, ixHospitalization, 112–13Humanity, 38–39, 52, 100, 156Hume, David, 164, 178Humility, 232

Hydration, 128

Ideals (moral), 40, 44, 167, 169–70, 177, 191,197, 255. See also Virtue(s) and virtuetheory

Impartiality, 109, 134, 139, 154, 161, 166, 169,203–04

Informed consentaccepted by the AMA, 58

and Anglo-American law, 54–56, 66–68competence to make decisions, 56, 58,

61–65, 72–75, 135compromised autonomy or consent, 24,

37, 70, 80, 137, 142–44the concept and elements of, x, xvii,59–65, 218–20, 239–40

its connection to autonomy, 9, 21, 61–65,72–75, 82, 88

demanded by Nuremberg MilitaryTribunals, 56

and disclosure, 52, 60, 63

I N D E X 285

Informed consent (Continued )Dworkin’s theories of, x, xxiv, 89, 219,

240–41, 250–51exceptions to the rules of, 64, 73–75, 274–75history of, xvi–xvii, 50–59, 180and intention (intentional action),

63–66, 71–72, 80, 83–85, 270the justification for not obtaining, 73–75,

109–11, 274–75meaning(s) of, 61–64, 239–40Onora O’Neill on, 81

in pharmaceutical trials, 133ff, 142–44as a prima facie duty to obtain, 74–75and understanding, 37, 60, 63–65, 68,

83–86, 88, 93, 251–52and voluntariness, 63–65, 70, 86–88, 90,

92, 137, 145, 219and vulnerable subjects and compliant

patients, 69–72See also Autonomy; Competence and

incompetence; Therapeuticmisconception; Vulnerabilites

Injustice, xix, 32, 139–40, 144, 217. See alsoJustice

Insanity, 39

Integrity (moral and personal), 155, 167,176, 203, 237–38, 265. See also Virtue(s)and virtue theory

Intelligence, 166. See also Moral status;Understanding

Intention (intentional action), 63–66,71–72, 80, 83–85, 89, 92, 94–95, 104–05,121, 123, 125, 138, 270, 275. See alsoInformed consent

Intuitions, moral, 27, 161, 213, 225, 234IRBs (Institutional Review Boards), 5,

22–23, 30. See also Research

Jehovah’s Witness’s refusal of bloodtransfusion, 81–82

Jewish Chronic Disease Hospital case, 20,25, 56–57

Jonsen, Albert, ix, xx, 3, 8, 10–12, 28, 163–64,231, 233

Justicebackground justice of pharmaceutical

trials, 133ff, 142–45

the concept of, 41

and the economically disadvantaged, xi,xix–xx, 4, 22, 132–46

formal and material principles of, 41

no single principle of, 41–42, 154as redress for disadvantaging conditions,132ff, 140, 144–45

and undue profits, xvii, xx, 137–41, 243for vulnerable groups, 4, 9, 21, 23,

25–27, 42See also Injustice; Principlism andprinciples (in bioethics);Vulnerabilities; Vulnerable personsand groups

Justificationcoherence as method of justification, xvi,45–47, 159, 183–85

and common-morality theory. SeeCommon morality

empirical justification (in the studyof ethics), xxii, 47, 55–56, 59–60,146, 178–81, 183, 186, 191, 194,200–04, 249

and method in biomedical ethics, 215–17,229, 235–39

normative justifications of commonmorality, 43, 180–82

normative conceptual justification, xii,xxii, xxiv, 191, 194–200, 204

for not obtaining consent, 73–75,109–11

of paternalism, 73–75, 101–04, 106–07,109–15, 117, 122–24

pragmatic justification, 179, 192–94,199–200

reflective equilibrium as a form of, xxiv,47, 159, 182–83, 189, 193–94, 225, 230,232–33, 236–37

utilitarian justifications of research, 7,25–26, 274–75

Kahn, Jeffrey, xiiKant, Immanuel, 8, 95, 192, 215–17,

254

Kantian theory, xxiv, 24, 161, 172, 215–17,221, 225, 230

Katz, Jay, ix–x, 5, 75, 219

286 I N D E X

Killing and letting die, xviii–xix, xxiv,120–30, 230, 241–42

Kindness, 39, 165, 176King, Nancy, xKing, Patricia, ixKukla, Rebecca, ix

Langdell, Christopher Columbus, 222

Lawthe case method in, xiii, 212, 221–26of disclosure and liability, xix, 61, 66–68,

73, 114, 215and informed consent, 54–56, 58, 60–63,

66–68, 142–44, 157–58, 219–20Negligence, 55, 66, 121, 277

Lederer, Susan, xiiLevine, Robert, ix–xiLiability, xix, 61, 66, 68, 73, 114, 215Liberty-limiting principles (autonomy-

limiting principles), 102–06, 120, 129.See also Autonomy

Lie, Reidar, x, xvLife-sustaining technologies, 81, 121–22, 125,

128, 242Lindsay, Ronald, ix, xiLipsett, Mortimer, 69

Locke, John, 195–96, 249Love, 39, 84, 86, 95–96, 253–54Lying, 52, 54, 91, 177, 198, 219

MacIntyre, Alasdair, 217–18, 237Mackie, John, 157

Manipulation, 19, 69–71, 86–87, 93, 108,136–37, 143

Marquis, Donald, xiMarshall, Ernest, ixMastectomies, 55, 115. See also Breast

implants (and the FDA)Mastroianni, Anna, xiiMcCarthy, Charles, ixMedical education, 184

Meisel, Alan, x, 219Mill, John Stuart, 8, 102–03, 109–12, 115–17,

156, 192, 216Miller, Frank, ix–x, xv, xvii, 237–38Minors, 64, 73Mishkin, Barbara, ix, 8

Moor, James (Jim), xi, xxiMoral change, 175, 185–86, 200Moral excellence. See Ideals (moral);

Virtue(s) and virtue theoryMoral rules, 27, 154, 156, 160, 168–70, 186,

190–93, 199–200, 218–21, 234–35Moral statusconcept of, xvii, 79, 186and person theory, xxiv, 25, 79, 93–97,186, 248–50, 252–57

Morality. See Common MoralityMoreno, Jonathan, xii, 239–40Motivesand autonomy and intention, 84, 87,90–92, 96, 108, 134, 239, 247

and person theory, 249–53, 256and virtue theory, 165–66

Murder, 125

National Commission for the Protection ofHuman Subjects of Biomedical andBehavioral Research, ix, xiii–xv, 4–11,18–30, 36, 69–70, 116–17

National security exceptions to unethicaltreatment, 274–75

Natural law, 194–95Negligence, 55, 66, 121, 277Nonmaleficenceand balancing benefits and harms, 22, 35,

39, 127beneficence contrasted to, 38–40,

169–70, 197concept of, 38–39duties or obligations of, 38–40and killing/letting die, 120–30, 241lessening harm as the goal of morality,192, 196–98, 201

and not causing harm, 11, 22, 36, 38–39,51, 73–74, 101, 127, 136, 154, 156, 169, 192,216, 254, 263, 266, 271–73

paternalism and prevention orreduction of harm, 101–04, 106–07,110–15, 122–24, 242

and preventing or removing harms, 7,40, 43, 109, 113, 127, 138, 176

See also Beneficence; Harms; Killing andLetting die; Nonmalevolence

I N D E X 287

Nonmalevolence, 176

Nozick, Robert, 217, 219Nuremberg trials and Code, 30, 56,

69, 273

Objectives (goals) of morality, xxi, 42–43,169, 171, 175–81, 185, 187, 192–93. See alsoCommon morality

O’Neill, Onora, 81, 217

Paincapacity to feel, 247, 253–54and distress, 114, 128, 134, 136, 141–42, 235extreme, 111, 125obligation not to cause, 169, 176, 191,

196, 255in subjects of research, 142

See also Distress; Harms;Nonmaleficence; Palliative care;Physician-hastened death; Suicide

Palliative care, 128

Paradigms and paradigm cases. SeeCasuistry; Principlism and principles(in bioethics)

Partiality. See ImpartialityPaternalismantipaternalist critique of, 109–11as an autonomy-limiting principle,

102–03beneficence and paternalism, 52, 101,

103–06, 110–17best interest(s) and, 103, 105, 112concept (and definition) of, 103–06government, 115–17in involuntary hospitalization, 112–13justification of, 73–75, 109–11, 117paternalistic intent vs. paternalistic

effect, 107–08paternalistic overprotection, 115, 126,

132, 142and prevention or reduction of harms,

101–04, 106–07, 110–15in public policies, 107–08, 110–11, 115–17in research ethics, 72, 116–17, 136, 142in some FDA policies and decisions, 115

in suicide intervention, 112

in treatment refusal cases, 113–14

weak (soft) and strong (hard), 106–07in withholding information, 50, 70, 73,

87, 104, 114–15, 117, 137, 219See also Autonomy; Beneficence;Nonmaleficence

Pellegrino, Edmund, ix–xi, xx, 167Percival, Thomas, 38, 50–52, 156, 273Permission, 21, 53, 60–61, 66, 74, 143, 265,

274. See also Informed consentPersons and theory of personsbrain-damaged persons, 96, 253connection to rights theory, 248,254–55

degrees of personhood, 252

disabled and handicapped persons, 7, 69,105, 134, 204, 255

incompetent persons, 9, 22, 24–25, 62,73, 80

Locke’s analysis of, 249

metaphysical personhood, xxiv, 94–96,247–52

moral personhood, xxiv, 94–96, 248,252–53

moral status in the absence ofpersonhood, 25, 253–54

respect for persons (principle of), 4, 7–9,13, 21, 24–27, 93–94, 214, 240, 250

and the theory of autonomy, 79, 89, 93ffTooley on, 248–50vagueness in the concept of persons, xxv,255–57

Persuasion, 70–71, 86, 137Pharmaceutical industryaccusations of unethical conduct in the,xi, xix–xx, 132

background justice and the, 144–45payment schemes for research subjects inthe, 137–42

and problems of exploitation of researchsubjects, 132–46

undue profit and undue influence in the,137–42

vulnerability of research subjects and the,133–37

Physician-hastened death and physician-assisted suicide, xi, xix, 120–29, 184,238, 241. See also Beneficence; Killing

288 I N D E X

and letting die; Slippery-slopearguments; Suicide

Placebo-controlled trials, 185. See alsoClinical trials; Research

Pluralism and relativism (moral), 171,177–78, 195, 231–35

Plutonium injections and experimentation,264–70

Pogge, Thomas, xiPoverty, 139. See also Economically

disadvantaged subjects; JusticePowers, Madison, xiPragmatic justification, 179, 192–94,

199–200Prejudice, 164, 213. See also Bias;

DiscriminationPresident’s Commission for the Study of

Ethical Problems in Medicine andBiomedical and Behavioral Research,58–60, 69–70, 148–49

Prevention, 117

Prima facie obligations and rights, xvi, 41,44–47, 103, 168, 242, 250. See alsoBalancing moral norms; Specification

Primum non nocere, 156. See also Harms;Nonmaleficence

Principlism and principles (in bioethics)ACHRE’s framework of principles, xxvi,263ff

and alternative paradigms, 153, 159ff, 172,176

balancing principles, xviii, 27, 103–06,110–12, 117, 169–70

balancing benefits, costs, and risks, 22,36, 39–40

Clouser on, 218–220and the common morality, 43–44, 176,

189fffour principles framework, ix–xiii, xvi,xviii, xx–xxi, xxiv, 10, 35–42, 47, 154–57,167–71, 189ff

hierarchical ranking of principles,28, 40

historical origins of, 3ff, 35–36, 153,155–56

the specification of principles, 45–47,157–59, 214ff

in the work of the NationalCommission for the Protection ofHuman Subjects of Biomedicaland Behavioral Research, ix,xiii–xv, 4–11, 18–30, 36, 69–70,116–17

See also PaternalismPrisoners as research subjects, 19, 58,

69–72, 116–17, 134–36

Quality of life, 185

Quante, Michael, xiiQuill, Timothy, xi, xviiiQuinlan, Karen Ann, xix

Rachels, James, xiRationality, xxiv, 85, 93–95, 108, 112, 190–91,

201, 248–52Rauprich, Oliver, ix, 202Rawls, John, 156, 159, 189, 193, 216–17,

225, 236Reciprocity, 6, 252Reflective equilibrium, xxiv, 47, 159,

182–83, 189, 193–94, 225, 230, 232–33,236–37

Refusal of treatment, xix, 37, 50,53, 62–63, 65, 72–74, 81–82,85, 104, 113ff, 122–23, 125, 127,241–42. See also Informed consent;Jehovah’s Witness’s refusal of bloodtransfusion

Regulations, 5, 13, 19, 26, 28–30, 59, 104,116, 238

Relativism. See Pluralism and relativism(moral)

ResearchAdvisory Committee on the HumanRadiation Experiments, xii, xxv, 29,261–78

access to clinical trials and results ofresearch, 26, 36

the Belmont Report on research ethics,9, 21, 23–26

conducted in clinical trials, xix, 26, 55,69–70, 132, 136, 138, 141

IRBs (Institutional Review Boards),22–23, 30

I N D E X 289

Research (Continued )National Commission for the Protection

of Human Subjects of Biomedical andBehavioral Research, ix, xiii–xv, 4–11,18–30, 36, 69–70, 116–17

paternalism in research ethics, 72, 116–17,136, 142

payment schemes for research subjects,137–42

President’s Commission for the Study ofEthical Problems in Medicine andBiomedical and Behavioral Researchon, 58–60, 69–70, 148–49

prisoners as research subjects, 19, 58,69–72, 116–17, 134–36

rights of research subjects, 13, 57, 134–35,144, 269ff

scandals that motivated development ofresearch ethics, 19–21, 56–57

utilitarian justifications of, 7, 25–26,274–75

See alsoAnimals (nonhuman); Autonomy;Double-blind studies; Experimentation;Informed consent; IRBs; Placebo-controlled trials; Therapeuticmisconception; Tuskegee SyphilisStudy; Vulnerable persons and groups

Respect for autonomy (principle of). SeeAutonomy; Principlism and principles(in bioethics)

Respect for persons (principle of), 4, 7–9,13, 21, 24–27, 93–94, 214, 240, 250

Retrospective moral judgments, xxv–xxvi,262–78

Richardson, Henry, x, xii, 182Rightsof animals, 253–56, 260autonomy-based, xvii, xix, 37, 65, 79ff,

102, 157, 232connection to persons and theories of

persons, xxv, 247–48, 254–55correlativity with obligations, 255

to health care, 157

legal and political, 41, 54, 58, 116prima facie obligations and rights,

xvi, 41, 44–47, 103, 113, 115, 168, 242,250, 275

of research subjects, 13, 57, 134–35, 144, 269ffSee also Autonomy; Balancing moralnorms; Specification

Risk. See Balancing moral norms;Beneficence; Harms

Robertson, John, ixRole responsibilities in professions, xxiii,

42, 56, 101, 156, 158, 169–70, 237, 272Ross, W.D., 6, 23, 44Rothman, David, 30

Ryan, Kenneth, ix

Sacrifice (personal), 40, 274Safeguards, 22, 70, 134, 238Secrecy, 214. See also ConfidentialitySeldin, Donald, ixSelf-defense, 126

Self-determination, 24–25, 51, 54, 73, 79, 157,263. See also Autonomy

Self-governance. See AutonomySinger, Peter, 230

Slavery, 186

Slippery-slope arguments, 184

Smith, David, xSocrates, 211, 213, 224–26Specification, x, xvi, 27, 41, 43–47, 155–63,

182–85, 199, 201, 215. See alsoPrinciplism and principles (inbioethics)

Standing (moral). See Moral statusStereotypes, 135

Strong, Carson, ix–xi, xiv, xviii, xx, 70Suffering. See also Distress; Harms;

Nonmaleficence; Pain; Physician-hastened death; Suicide

Suicide, xix, 104, 107–14, 117, 128, 130, 184,238, 241. See alsoKilling and letting die;Physician-hastened death

Sunstein, Cass, 108

Supererogation, See Ideals (moral)Surrogate decision makersauthority, rules of, 44, 73, 117, 154guardians as, 73, 81, 104, 115, 255

Taylor, James Stacey, x, xviiTerminal sedation, 128, 241Therapeutic misconception, 86

290 I N D E X

Therapeutic privilege, 73–74Toulmin, Stephen, ix–x, xx, 3–4, 10–11, 15,

28, 231, 233Truth-telling and truthfulness. See

Disclosure; Honesty; Informedconsent

Turner, Leigh, xxi, 175, 177–80Tuskegee Syphilis Study, 19–20, 25,

57, 71

Uncertainty, 19, 52, 65, 240Understanding. See Informed consentUndue influence, 116, 132, 137–43, 213Undue profit, xvii, xx, 137–43. See also

ExploitationUniversal morality and universal principles,

xxi, 6, 23–24, 28, 42–43, 161, 164, 171,175ff, 190, 196, 200, 202, 231–35, 263,271–72, 275. See also Commonmorality; Principlism and principles(in bioethics)

Utilitarianismbalancing moral norms in, 25–26,

110–11Bentham and, 253

in the justification of research, 7, 25–27Mill and, 8, 102–03, 109–12, 115–17, 156,192, 216

and practical ethics, xxiv, 230and sacrifice, 40, 274

Veatch, Robert, ix–x, xii, xvi, 24–25Veracity, 24, 51–52, 214. See also Disclosure;

Informed ConsentVice, 64, 176, 185, 198, 255. See also Virtue(s)

and virtue theoryVirtue(s) and virtue theoryand autonomy theory, 80

benevolence as a virtue, 39

care and caring as fundamental virtues,165–66

conscientiousness as a virtue, 167

ideals and, 40, 44, 167, 169–70, 177, 191,197, 255

integrity as a virtue, 155, 167, 176, 203,237–38, 265. See also Integrity (moraland personal)

motives and their relation to virtue,165–66

moral traits of character, 165–66, 176Pellegrino’s theory of, 167

in professional roles, 42, 101veracity as a virtue. See Veracitythe virtue of discernment, 160, 166virtue theory as an alternative toprinciplism, xx, 153, 156, 164–67,171, 230, 232

See also ViceVolition, 89–94, 251Voluntariness, 21, 37, 63–65, 70, 86–88, 90,

92, 137, 145, 219. See also Informedconsent

Vulnerabilities, 35, 133–35, 142, 145, 193,198, 202

Vulnerable persons and groupscategories of, 134–35economically disadvantaged participantsin research as, xi, xix, 4, 132–41

and informed consent involvingcompliant patients, 69–72

justice concerns about, 4, 9, 21, 23, 25–27,42, 134, 136, 139–41

See also Exploitation; Research

Wantons, 91–92Waiver(s), 28, 73Walters, LeRoy, ixWarnock, G. J., 192–93, 196Wendler, David, x, xvWikler, Daniel, xi, 219Withdrawal and withholding of medical

technology, 113–14, 121–25, 128. See alsoKilling and letting die; Refusal oftreatment

Withholding information, 50, 70, 73, 87,104, 114, 117, 137, 219. See alsoDisclosure; Informed consent;Veracity

World Medical Association (WMA), 130

Yesley, Michael, ix, 4–9

Zealots (moral and religious),82, 181

I N D E X 291

Standing on Principles Collected Essays

Documents

Transcript of Standing on Principles Collected Essays